Token | , |
Token ID | 11 |
Feature activation | -9.35e-01 |
Pos logit contributions | |
done | +0.026 |
wrong | +0.025 |
sky | +0.025 |
missing | +0.024 |
made | +0.023 |
Neg logit contributions | |
cess | -0.056 |
charged | -0.052 |
abbit | -0.050 |
ck | -0.050 |
aret | -0.049 |
Token | Bugs |
Token ID | 44991 |
Feature activation | -6.38e-01 |
Pos logit contributions | |
done | +0.027 |
wrong | +0.025 |
missing | +0.025 |
has | +0.024 |
sky | +0.024 |
Neg logit contributions | |
cess | -0.015 |
chick | -0.013 |
ck | -0.013 |
chicks | -0.012 |
abbit | -0.012 |
Token | said |
Token ID | 531 |
Feature activation | -3.64e+00 |
Pos logit contributions | |
wrong | +0.009 |
done | +0.009 |
missing | +0.008 |
sky | +0.008 |
can | +0.008 |
Neg logit contributions | |
cess | -0.020 |
charged | -0.018 |
ck | -0.018 |
abbit | -0.018 |
aret | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | -1.23e+00 |
Pos logit contributions | |
done | +0.049 |
sky | +0.046 |
wrong | +0.045 |
missing | +0.044 |
made | +0.044 |
Neg logit contributions | |
cess | -0.115 |
charged | -0.103 |
abbit | -0.103 |
ck | -0.102 |
oxy | -0.101 |
Token | " |
Token ID | 366 |
Feature activation | +1.75e+00 |
Pos logit contributions | |
done | +0.017 |
wrong | +0.016 |
sky | +0.015 |
made | +0.015 |
missing | +0.015 |
Neg logit contributions | |
cess | -0.039 |
abbit | -0.035 |
ck | -0.035 |
charged | -0.035 |
aret | -0.035 |
Token | Twe |
Token ID | 32665 |
Feature activation | +8.45e+00 |
Pos logit contributions | |
cess | +0.050 |
charged | +0.045 |
aret | +0.044 |
ck | +0.044 |
abbit | +0.043 |
Neg logit contributions | |
made | -0.030 |
wrong | -0.030 |
sky | -0.030 |
had | -0.029 |
can | -0.029 |
Token | ety |
Token ID | 2963 |
Feature activation | +2.00e+00 |
Pos logit contributions | |
cess | +0.183 |
rily | +0.165 |
aret | +0.163 |
keys | +0.156 |
rums | +0.155 |
Neg logit contributions | |
missing | -0.182 |
made | -0.180 |
done | -0.179 |
wrong | -0.178 |
had | -0.178 |
Token | , |
Token ID | 11 |
Feature activation | +1.33e-01 |
Pos logit contributions | |
cess | +0.068 |
charged | +0.063 |
ck | +0.062 |
aret | +0.061 |
abbit | +0.061 |
Neg logit contributions | |
done | -0.022 |
wrong | -0.022 |
sky | -0.021 |
made | -0.021 |
missing | -0.021 |
Token | I |
Token ID | 314 |
Feature activation | -7.52e-01 |
Pos logit contributions | |
cess | +0.003 |
chick | +0.003 |
swat | +0.003 |
ck | +0.003 |
God | +0.003 |
Neg logit contributions | |
sky | -0.002 |
done | -0.002 |
boards | -0.002 |
missing | -0.002 |
wrong | -0.002 |
Token | have |
Token ID | 423 |
Feature activation | -2.82e+00 |
Pos logit contributions | |
sky | +0.011 |
wrong | +0.011 |
boards | +0.010 |
done | +0.010 |
missing | +0.010 |
Neg logit contributions | |
cess | -0.022 |
ck | -0.022 |
charged | -0.021 |
oxy | -0.020 |
chick | -0.020 |
Token | a |
Token ID | 257 |
Feature activation | -1.16e-01 |
Pos logit contributions | |
done | +0.034 |
wrong | +0.033 |
sky | +0.033 |
can | +0.032 |
boards | +0.031 |
Neg logit contributions | |
cess | -0.084 |
charged | -0.080 |
oxy | -0.079 |
chick | -0.079 |
chicks | -0.078 |
Token | ety |
Token ID | 2963 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
cess | +0.192 |
rily | +0.181 |
wrists | +0.179 |
aret | +0.174 |
rums | +0.171 |
Neg logit contributions | |
made | -0.143 |
can | -0.142 |
missing | -0.141 |
m | -0.140 |
| -0.139 |
Token | . |
Token ID | 13 |
Feature activation | +5.48e-01 |
Pos logit contributions | |
cess | +0.080 |
charged | +0.073 |
aret | +0.073 |
ck | +0.072 |
oxy | +0.072 |
Neg logit contributions | |
wrong | -0.039 |
done | -0.037 |
missing | -0.037 |
made | -0.036 |
on | -0.036 |
Token | " |
Token ID | 366 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
cess | +0.011 |
charged | +0.009 |
ck | +0.009 |
aret | +0.009 |
abbit | +0.009 |
Neg logit contributions | |
done | -0.014 |
wrong | -0.014 |
made | -0.013 |
sky | -0.013 |
missing | -0.013 |
Token | Hello |
Token ID | 15496 |
Feature activation | +1.50e+00 |
Pos logit contributions | |
cess | +0.076 |
charged | +0.070 |
aret | +0.069 |
abbit | +0.067 |
ck | +0.067 |
Neg logit contributions | |
made | -0.026 |
sky | -0.026 |
had | -0.025 |
wrong | -0.025 |
can | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +2.08e+00 |
Pos logit contributions | |
cess | +0.046 |
charged | +0.042 |
ck | +0.042 |
aret | +0.042 |
abbit | +0.041 |
Neg logit contributions | |
wrong | -0.022 |
done | -0.021 |
made | -0.020 |
sky | -0.020 |
missing | -0.020 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
cess | +0.041 |
ck | +0.032 |
chick | +0.032 |
abbit | +0.032 |
charged | +0.031 |
Neg logit contributions | |
done | -0.057 |
wrong | -0.053 |
missing | -0.052 |
boards | -0.051 |
made | -0.051 |
Token | ety |
Token ID | 2963 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
cess | +0.183 |
rily | +0.180 |
wrists | +0.179 |
ishment | +0.169 |
aret | +0.168 |
Neg logit contributions | |
can | -0.172 |
made | -0.171 |
had | -0.168 |
has | -0.168 |
sky | -0.165 |
Token | . |
Token ID | 13 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
cess | +0.128 |
ishment | +0.122 |
wrists | +0.121 |
charged | +0.121 |
oxy | +0.120 |
Neg logit contributions | |
wrong | -0.063 |
made | -0.061 |
had | -0.060 |
on | -0.060 |
can | -0.058 |
Token | You |
Token ID | 921 |
Feature activation | -9.32e-01 |
Pos logit contributions | |
cess | +0.021 |
charged | +0.019 |
abbit | +0.018 |
ck | +0.018 |
aret | +0.018 |
Neg logit contributions | |
done | -0.012 |
wrong | -0.012 |
made | -0.011 |
missing | -0.011 |
sky | -0.011 |
Token | sing |
Token ID | 1702 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
wrong | +0.009 |
boards | +0.009 |
missing | +0.009 |
sky | +0.008 |
done | +0.008 |
Neg logit contributions | |
cess | -0.031 |
charged | -0.029 |
ck | -0.029 |
chick | -0.027 |
ymes | -0.027 |
Token | so |
Token ID | 523 |
Feature activation | +7.61e-01 |
Pos logit contributions | |
cess | +0.019 |
charged | +0.018 |
ck | +0.018 |
aret | +0.018 |
abbit | +0.017 |
Neg logit contributions | |
done | -0.007 |
wrong | -0.007 |
missing | -0.006 |
sky | -0.006 |
made | -0.006 |
Token | a |
Token ID | 257 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
wrong | +0.005 |
done | +0.005 |
made | +0.005 |
sky | +0.005 |
on | +0.004 |
Neg logit contributions | |
cess | -0.035 |
charged | -0.032 |
aret | -0.032 |
ck | -0.031 |
abbit | -0.031 |
Token | little |
Token ID | 1310 |
Feature activation | +4.11e+00 |
Pos logit contributions | |
cess | +0.057 |
charged | +0.056 |
rums | +0.056 |
ck | +0.054 |
eter | +0.054 |
Neg logit contributions | |
big | -0.019 |
had | -0.018 |
can | -0.017 |
wrong | -0.017 |
made | -0.017 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
cess | +0.126 |
ck | +0.114 |
charged | +0.114 |
rums | +0.112 |
aret | +0.110 |
Neg logit contributions | |
sky | -0.068 |
can | -0.067 |
made | -0.066 |
wrong | -0.065 |
on | -0.064 |
Token | ie |
Token ID | 494 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
cess | +0.114 |
ishment | +0.108 |
eter | +0.105 |
rily | +0.102 |
charged | +0.102 |
Neg logit contributions | |
made | -0.076 |
on | -0.074 |
had | -0.070 |
all | -0.070 |
wrong | -0.069 |
Token | named |
Token ID | 3706 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
cess | +0.137 |
ishment | +0.136 |
eter | +0.132 |
rums | +0.129 |
rily | +0.128 |
Neg logit contributions | |
made | -0.058 |
had | -0.056 |
on | -0.056 |
all | -0.052 |
wrong | -0.049 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.41e+00 |
Pos logit contributions | |
cess | +0.078 |
ymes | +0.072 |
charged | +0.071 |
oing | +0.069 |
rums | +0.069 |
Neg logit contributions | |
all | -0.055 |
wrong | -0.054 |
made | -0.052 |
sky | -0.052 |
missing | -0.051 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
cess | +0.229 |
rums | +0.204 |
keys | +0.198 |
wrists | +0.198 |
rily | +0.194 |
Neg logit contributions | |
m | -0.150 |
missing | -0.148 |
had | -0.141 |
made | -0.137 |
wrong | -0.131 |
Token | . |
Token ID | 13 |
Feature activation | -3.70e-01 |
Pos logit contributions | |
cess | +0.133 |
ishment | +0.126 |
eter | +0.124 |
rily | +0.122 |
charged | +0.121 |
Neg logit contributions | |
all | -0.039 |
made | -0.039 |
had | -0.038 |
on | -0.036 |
wrong | -0.036 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.12e+00 |
Pos logit contributions | |
done | +0.012 |
wrong | +0.012 |
made | +0.011 |
missing | +0.011 |
can | +0.011 |
Neg logit contributions | |
cess | -0.005 |
charged | -0.004 |
ck | -0.004 |
aret | -0.004 |
abbit | -0.004 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
cess | +0.173 |
keys | +0.147 |
wrists | +0.147 |
commands | +0.143 |
tune | +0.142 |
Neg logit contributions | |
had | -0.146 |
missing | -0.145 |
made | -0.143 |
done | -0.139 |
wrong | -0.139 |
Token | lived |
Token ID | 5615 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
cess | +0.049 |
charged | +0.046 |
ck | +0.046 |
aret | +0.045 |
abbit | +0.045 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
sky | -0.012 |
missing | -0.012 |
made | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
wrong | +0.003 |
done | +0.003 |
made | +0.003 |
on | +0.003 |
sky | +0.003 |
Neg logit contributions | |
cess | -0.027 |
charged | -0.025 |
aret | -0.025 |
ck | -0.024 |
oing | -0.024 |
Token | little |
Token ID | 1310 |
Feature activation | +4.07e+00 |
Pos logit contributions | |
cess | +0.066 |
charged | +0.064 |
rums | +0.064 |
ck | +0.062 |
eter | +0.062 |
Neg logit contributions | |
big | -0.023 |
wrong | -0.023 |
had | -0.023 |
can | -0.022 |
made | -0.022 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
cess | +0.122 |
charged | +0.110 |
ck | +0.110 |
rums | +0.107 |
aret | +0.107 |
Neg logit contributions | |
sky | -0.069 |
can | -0.068 |
made | -0.067 |
wrong | -0.067 |
on | -0.065 |
Token | ie |
Token ID | 494 |
Feature activation | +4.05e+00 |
Pos logit contributions | |
cess | +0.118 |
ishment | +0.111 |
eter | +0.110 |
rily | +0.107 |
charged | +0.106 |
Neg logit contributions | |
made | -0.073 |
on | -0.071 |
had | -0.067 |
all | -0.066 |
wrong | -0.066 |
Token | named |
Token ID | 3706 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
cess | +0.135 |
ishment | +0.132 |
eter | +0.132 |
rums | +0.126 |
rily | +0.125 |
Neg logit contributions | |
made | -0.061 |
had | -0.060 |
on | -0.059 |
all | -0.055 |
wrong | -0.053 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.40e+00 |
Pos logit contributions | |
cess | +0.081 |
ymes | +0.077 |
charged | +0.076 |
rums | +0.073 |
aret | +0.073 |
Neg logit contributions | |
all | -0.051 |
wrong | -0.050 |
made | -0.049 |
sky | -0.049 |
missing | -0.048 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
cess | +0.228 |
rums | +0.199 |
wrists | +0.191 |
commands | +0.191 |
tune | +0.190 |
Neg logit contributions | |
missing | -0.149 |
m | -0.145 |
had | -0.142 |
made | -0.137 |
wrong | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
cess | +0.132 |
ishment | +0.124 |
eter | +0.123 |
rily | +0.120 |
charged | +0.120 |
Neg logit contributions | |
made | -0.038 |
had | -0.038 |
all | -0.037 |
wrong | -0.036 |
on | -0.036 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.19e+00 |
Pos logit contributions | |
done | +0.012 |
wrong | +0.012 |
made | +0.011 |
missing | +0.011 |
can | +0.011 |
Neg logit contributions | |
cess | -0.005 |
charged | -0.004 |
ck | -0.003 |
abbit | -0.003 |
aret | -0.003 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
cess | +0.172 |
wrists | +0.143 |
tune | +0.142 |
commands | +0.141 |
keys | +0.141 |
Neg logit contributions | |
missing | -0.151 |
had | -0.151 |
made | -0.148 |
wrong | -0.145 |
sky | -0.144 |
Token | was |
Token ID | 373 |
Feature activation | -1.06e+00 |
Pos logit contributions | |
cess | +0.048 |
ck | +0.045 |
charged | +0.045 |
aret | +0.044 |
abbit | +0.044 |
Neg logit contributions | |
wrong | -0.015 |
done | -0.015 |
missing | -0.013 |
sky | -0.013 |
made | -0.013 |
Token | and |
Token ID | 290 |
Feature activation | +1.66e+00 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.013 |
ck | +0.013 |
aret | +0.013 |
abbit | +0.013 |
Neg logit contributions | |
done | -0.008 |
wrong | -0.008 |
made | -0.007 |
sky | -0.007 |
missing | -0.007 |
Token | said |
Token ID | 531 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
cess | +0.050 |
charged | +0.046 |
ck | +0.045 |
aret | +0.045 |
abbit | +0.044 |
Neg logit contributions | |
wrong | -0.025 |
done | -0.025 |
made | -0.024 |
sky | -0.024 |
missing | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | -1.03e-01 |
Pos logit contributions | |
wrong | +0.027 |
done | +0.027 |
made | +0.025 |
sky | +0.024 |
missing | +0.024 |
Neg logit contributions | |
cess | -0.087 |
charged | -0.081 |
ck | -0.080 |
aret | -0.079 |
abbit | -0.079 |
Token | " |
Token ID | 366 |
Feature activation | +2.09e+00 |
Pos logit contributions | |
done | +0.001 |
wrong | +0.001 |
made | +0.001 |
sky | +0.001 |
missing | +0.001 |
Neg logit contributions | |
cess | -0.003 |
charged | -0.003 |
ck | -0.003 |
aret | -0.003 |
abbit | -0.003 |
Token | Hi |
Token ID | 17250 |
Feature activation | +8.35e-01 |
Pos logit contributions | |
cess | +0.069 |
charged | +0.064 |
ck | +0.064 |
aret | +0.063 |
abbit | +0.062 |
Neg logit contributions | |
wrong | -0.026 |
sky | -0.025 |
made | -0.025 |
done | -0.025 |
had | -0.024 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.39e+00 |
Pos logit contributions | |
cess | +0.019 |
charged | +0.017 |
ck | +0.017 |
aret | +0.017 |
oxy | +0.016 |
Neg logit contributions | |
wrong | -0.019 |
done | -0.018 |
made | -0.018 |
sky | -0.018 |
can | -0.018 |
Token | ety |
Token ID | 2963 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
cess | +0.192 |
rily | +0.178 |
wrists | +0.176 |
rums | +0.175 |
tune | +0.171 |
Neg logit contributions | |
m | -0.162 |
had | -0.162 |
can | -0.161 |
missing | -0.160 |
made | -0.160 |
Token | ! |
Token ID | 0 |
Feature activation | +3.34e-01 |
Pos logit contributions | |
cess | +0.147 |
ck | +0.144 |
charged | +0.143 |
ishment | +0.141 |
oxy | +0.140 |
Neg logit contributions | |
wrong | -0.042 |
on | -0.039 |
made | -0.039 |
had | -0.039 |
can | -0.039 |
Token | I |
Token ID | 314 |
Feature activation | -3.82e-01 |
Pos logit contributions | |
cess | +0.010 |
charged | +0.009 |
ck | +0.008 |
abbit | +0.008 |
aret | +0.008 |
Neg logit contributions | |
done | -0.006 |
wrong | -0.006 |
made | -0.005 |
missing | -0.005 |
sky | -0.005 |
Token | am |
Token ID | 716 |
Feature activation | +6.45e-02 |
Pos logit contributions | |
boards | +0.005 |
wrong | +0.005 |
sky | +0.005 |
done | +0.004 |
best | +0.004 |
Neg logit contributions | |
ck | -0.012 |
cess | -0.011 |
oing | -0.011 |
charged | -0.011 |
chick | -0.011 |
Token | drinking |
Token ID | 7722 |
Feature activation | -5.93e-01 |
Pos logit contributions | |
cess | +0.002 |
ck | +0.002 |
charged | +0.002 |
abbit | +0.002 |
aret | +0.002 |
Neg logit contributions | |
done | -0.001 |
wrong | -0.001 |
sky | -0.001 |
made | -0.001 |
can | -0.001 |
Token | a |
Token ID | 257 |
Feature activation | +1.93e+00 |
Pos logit contributions | |
wrong | +0.004 |
done | +0.004 |
made | +0.004 |
sky | +0.003 |
on | +0.003 |
Neg logit contributions | |
cess | -0.028 |
charged | -0.026 |
aret | -0.026 |
ck | -0.026 |
abbit | -0.025 |
Token | little |
Token ID | 1310 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
cess | +0.066 |
charged | +0.064 |
rums | +0.063 |
ck | +0.062 |
aret | +0.061 |
Neg logit contributions | |
big | -0.023 |
wrong | -0.023 |
can | -0.022 |
had | -0.022 |
sky | -0.022 |
Token | bird |
Token ID | 6512 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
cess | +0.126 |
ck | +0.114 |
charged | +0.114 |
rums | +0.111 |
aret | +0.111 |
Neg logit contributions | |
sky | -0.070 |
can | -0.068 |
wrong | -0.068 |
made | -0.067 |
on | -0.065 |
Token | ie |
Token ID | 494 |
Feature activation | +4.15e+00 |
Pos logit contributions | |
cess | +0.121 |
ishment | +0.113 |
eter | +0.111 |
aret | +0.108 |
rily | +0.108 |
Neg logit contributions | |
made | -0.076 |
on | -0.075 |
had | -0.070 |
all | -0.070 |
wrong | -0.070 |
Token | named |
Token ID | 3706 |
Feature activation | +2.89e+00 |
Pos logit contributions | |
cess | +0.139 |
ishment | +0.136 |
eter | +0.134 |
rums | +0.131 |
aret | +0.129 |
Neg logit contributions | |
made | -0.060 |
on | -0.060 |
had | -0.058 |
all | -0.055 |
wrong | -0.053 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.37e+00 |
Pos logit contributions | |
cess | +0.084 |
ymes | +0.078 |
charged | +0.078 |
oing | +0.075 |
aret | +0.075 |
Neg logit contributions | |
all | -0.054 |
wrong | -0.053 |
made | -0.052 |
sky | -0.051 |
missing | -0.050 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.71e+00 |
Pos logit contributions | |
cess | +0.230 |
rums | +0.205 |
wrists | +0.199 |
rily | +0.198 |
tune | +0.195 |
Neg logit contributions | |
m | -0.146 |
missing | -0.144 |
had | -0.138 |
made | -0.134 |
wrong | -0.129 |
Token | . |
Token ID | 13 |
Feature activation | -3.97e-01 |
Pos logit contributions | |
cess | +0.138 |
ishment | +0.131 |
eter | +0.129 |
rily | +0.127 |
charged | +0.126 |
Neg logit contributions | |
all | -0.039 |
made | -0.038 |
had | -0.037 |
on | -0.037 |
wrong | -0.037 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
done | +0.013 |
wrong | +0.013 |
made | +0.012 |
missing | +0.012 |
can | +0.012 |
Neg logit contributions | |
cess | -0.005 |
charged | -0.004 |
ck | -0.004 |
abbit | -0.004 |
aret | -0.004 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
cess | +0.175 |
wrists | +0.150 |
tune | +0.147 |
keys | +0.146 |
commands | +0.145 |
Neg logit contributions | |
had | -0.139 |
missing | -0.137 |
made | -0.135 |
wrong | -0.132 |
done | -0.132 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.68e-01 |
Pos logit contributions | |
cess | +0.047 |
ck | +0.044 |
charged | +0.044 |
aret | +0.043 |
abbit | +0.042 |
Neg logit contributions | |
wrong | -0.015 |
done | -0.015 |
missing | -0.013 |
sky | -0.013 |
made | -0.013 |
Token | baby |
Token ID | 5156 |
Feature activation | +4.00e+00 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.013 |
ck | +0.013 |
aret | +0.013 |
oxy | +0.013 |
Neg logit contributions | |
wrong | -0.007 |
done | -0.007 |
best | -0.007 |
missing | -0.007 |
made | -0.007 |
Token | back |
Token ID | 736 |
Feature activation | +2.83e-01 |
Pos logit contributions | |
cess | +0.125 |
aret | +0.113 |
ck | +0.111 |
oxy | +0.110 |
charged | +0.110 |
Neg logit contributions | |
wrong | -0.063 |
made | -0.063 |
best | -0.062 |
on | -0.061 |
had | -0.061 |
Token | and |
Token ID | 290 |
Feature activation | +3.90e+00 |
Pos logit contributions | |
cess | +0.008 |
charged | +0.007 |
ck | +0.007 |
abbit | +0.007 |
aret | +0.007 |
Neg logit contributions | |
wrong | -0.005 |
done | -0.005 |
sky | -0.005 |
made | -0.004 |
missing | -0.004 |
Token | she |
Token ID | 673 |
Feature activation | -6.82e-01 |
Pos logit contributions | |
cess | +0.107 |
charged | +0.096 |
ck | +0.096 |
aret | +0.094 |
abbit | +0.092 |
Neg logit contributions | |
wrong | -0.070 |
done | -0.069 |
sky | -0.066 |
missing | -0.066 |
made | -0.065 |
Token | thanked |
Token ID | 26280 |
Feature activation | +2.83e+00 |
Pos logit contributions | |
wrong | +0.013 |
done | +0.013 |
made | +0.012 |
sky | +0.012 |
missing | +0.012 |
Neg logit contributions | |
cess | -0.018 |
charged | -0.017 |
ck | -0.016 |
aret | -0.016 |
abbit | -0.016 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.36e+00 |
Pos logit contributions | |
cess | +0.083 |
charged | +0.078 |
oing | +0.078 |
abbit | +0.077 |
ymes | +0.077 |
Neg logit contributions | |
all | -0.051 |
wrong | -0.046 |
on | -0.046 |
made | -0.046 |
had | -0.044 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
cess | +0.215 |
wrists | +0.189 |
tune | +0.185 |
tunes | +0.184 |
aret | +0.184 |
Neg logit contributions | |
m | -0.152 |
can | -0.143 |
had | -0.140 |
wrong | -0.138 |
made | -0.136 |
Token | and |
Token ID | 290 |
Feature activation | +3.35e+00 |
Pos logit contributions | |
cess | +0.160 |
charged | +0.152 |
ck | +0.151 |
oxy | +0.151 |
rums | +0.148 |
Neg logit contributions | |
wrong | -0.021 |
all | -0.021 |
on | -0.021 |
best | -0.019 |
done | -0.018 |
Token | the |
Token ID | 262 |
Feature activation | -1.25e-02 |
Pos logit contributions | |
cess | +0.097 |
charged | +0.090 |
abbit | +0.087 |
ck | +0.087 |
oxy | +0.086 |
Neg logit contributions | |
done | -0.056 |
made | -0.055 |
wrong | -0.054 |
had | -0.054 |
can | -0.052 |
Token | owl |
Token ID | 39610 |
Feature activation | +2.46e+00 |
Pos logit contributions | |
wrong | +0.000 |
done | +0.000 |
made | +0.000 |
sky | +0.000 |
missing | +0.000 |
Neg logit contributions | |
cess | -0.000 |
charged | -0.000 |
ck | -0.000 |
aret | -0.000 |
abbit | -0.000 |
Token | for |
Token ID | 329 |
Feature activation | +1.58e+00 |
Pos logit contributions | |
cess | +0.083 |
charged | +0.078 |
oxy | +0.077 |
aret | +0.077 |
ishment | +0.077 |
Neg logit contributions | |
on | -0.028 |
had | -0.028 |
wrong | -0.028 |
all | -0.027 |
made | -0.027 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.19e+00 |
Pos logit contributions | |
wrong | +0.034 |
done | +0.031 |
best | +0.028 |
missing | +0.027 |
sky | +0.027 |
Neg logit contributions | |
cess | -0.083 |
ck | -0.078 |
oxy | -0.077 |
abbit | -0.076 |
charged | -0.076 |
Token | a |
Token ID | 257 |
Feature activation | +3.58e-01 |
Pos logit contributions | |
done | +0.026 |
wrong | +0.024 |
can | +0.023 |
has | +0.022 |
missing | +0.022 |
Neg logit contributions | |
cess | -0.073 |
ck | -0.068 |
charged | -0.067 |
oxy | -0.066 |
chick | -0.066 |
Token | friendly |
Token ID | 8030 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
cess | +0.010 |
chick | +0.009 |
abbit | +0.009 |
oing | +0.009 |
tune | +0.009 |
Neg logit contributions | |
done | -0.006 |
wrong | -0.006 |
made | -0.005 |
missing | -0.005 |
boards | -0.005 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.84e+00 |
Pos logit contributions | |
cess | +0.061 |
charged | +0.054 |
ck | +0.053 |
aret | +0.053 |
abbit | +0.052 |
Neg logit contributions | |
wrong | -0.057 |
done | -0.057 |
made | -0.055 |
sky | -0.054 |
missing | -0.054 |
Token | named |
Token ID | 3706 |
Feature activation | +1.87e+00 |
Pos logit contributions | |
cess | +0.105 |
aret | +0.094 |
charged | +0.093 |
eter | +0.091 |
abbit | +0.091 |
Neg logit contributions | |
wrong | -0.072 |
made | -0.072 |
on | -0.071 |
had | -0.070 |
done | -0.069 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.35e+00 |
Pos logit contributions | |
cess | +0.042 |
charged | +0.041 |
ck | +0.040 |
aret | +0.039 |
eter | +0.038 |
Neg logit contributions | |
wrong | -0.042 |
made | -0.040 |
sky | -0.039 |
missing | -0.039 |
done | -0.039 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.10e+00 |
Pos logit contributions | |
cess | +0.185 |
rums | +0.170 |
aret | +0.165 |
wrists | +0.163 |
rily | +0.162 |
Neg logit contributions | |
missing | -0.172 |
m | -0.170 |
made | -0.166 |
| -0.164 |
had | -0.162 |
Token | who |
Token ID | 508 |
Feature activation | -2.05e+00 |
Pos logit contributions | |
cess | +0.097 |
charged | +0.088 |
aret | +0.087 |
ck | +0.086 |
oxy | +0.086 |
Neg logit contributions | |
wrong | -0.047 |
done | -0.044 |
made | -0.043 |
missing | -0.043 |
on | -0.043 |
Token | suggested |
Token ID | 5220 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
done | +0.018 |
wrong | +0.017 |
sky | +0.016 |
best | +0.015 |
missing | +0.014 |
Neg logit contributions | |
cess | -0.074 |
ck | -0.070 |
charged | -0.068 |
aret | -0.066 |
abbit | -0.066 |
Token | that |
Token ID | 326 |
Feature activation | -6.34e-01 |
Pos logit contributions | |
wrong | +0.024 |
done | +0.023 |
made | +0.022 |
sky | +0.021 |
missing | +0.021 |
Neg logit contributions | |
cess | -0.076 |
charged | -0.071 |
ck | -0.070 |
aret | -0.069 |
abbit | -0.069 |
Token | she |
Token ID | 673 |
Feature activation | +5.54e-01 |
Pos logit contributions | |
wrong | +0.015 |
done | +0.015 |
made | +0.014 |
sky | +0.014 |
missing | +0.014 |
Neg logit contributions | |
cess | -0.014 |
charged | -0.013 |
ck | -0.012 |
aret | -0.012 |
abbit | -0.012 |
Token | a |
Token ID | 257 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
done | +0.003 |
made | +0.003 |
wrong | +0.003 |
sky | +0.003 |
on | +0.003 |
Neg logit contributions | |
cess | -0.030 |
charged | -0.028 |
aret | -0.027 |
ck | -0.027 |
oing | -0.027 |
Token | little |
Token ID | 1310 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
cess | +0.060 |
charged | +0.059 |
rums | +0.058 |
ck | +0.057 |
aret | +0.056 |
Neg logit contributions | |
can | -0.022 |
had | -0.022 |
big | -0.022 |
sky | -0.021 |
wrong | -0.021 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.97e+00 |
Pos logit contributions | |
cess | +0.120 |
charged | +0.110 |
ck | +0.110 |
rums | +0.107 |
aret | +0.106 |
Neg logit contributions | |
sky | -0.067 |
can | -0.065 |
made | -0.065 |
wrong | -0.064 |
on | -0.063 |
Token | ie |
Token ID | 494 |
Feature activation | +3.94e+00 |
Pos logit contributions | |
cess | +0.120 |
ishment | +0.113 |
eter | +0.111 |
rily | +0.108 |
charged | +0.107 |
Neg logit contributions | |
made | -0.074 |
on | -0.073 |
had | -0.069 |
all | -0.068 |
wrong | -0.067 |
Token | named |
Token ID | 3706 |
Feature activation | +2.80e+00 |
Pos logit contributions | |
cess | +0.134 |
ishment | +0.131 |
eter | +0.129 |
rums | +0.125 |
rily | +0.124 |
Neg logit contributions | |
made | -0.056 |
had | -0.055 |
on | -0.054 |
all | -0.049 |
wrong | -0.049 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.27e+00 |
Pos logit contributions | |
cess | +0.079 |
ymes | +0.074 |
charged | +0.074 |
oing | +0.071 |
rums | +0.071 |
Neg logit contributions | |
all | -0.054 |
wrong | -0.052 |
made | -0.051 |
missing | -0.051 |
sky | -0.050 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.62e+00 |
Pos logit contributions | |
cess | +0.229 |
rums | +0.202 |
keys | +0.198 |
Amen | +0.194 |
rily | +0.194 |
Neg logit contributions | |
missing | -0.143 |
m | -0.143 |
had | -0.136 |
made | -0.134 |
wrong | -0.128 |
Token | . |
Token ID | 13 |
Feature activation | -3.38e-01 |
Pos logit contributions | |
cess | +0.134 |
ishment | +0.126 |
eter | +0.124 |
rily | +0.123 |
charged | +0.122 |
Neg logit contributions | |
made | -0.039 |
all | -0.038 |
had | -0.038 |
wrong | -0.037 |
on | -0.037 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.01e+00 |
Pos logit contributions | |
done | +0.011 |
wrong | +0.011 |
made | +0.010 |
missing | +0.010 |
can | +0.010 |
Neg logit contributions | |
cess | -0.004 |
charged | -0.003 |
ck | -0.003 |
aret | -0.003 |
abbit | -0.003 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
cess | +0.171 |
keys | +0.146 |
wrists | +0.143 |
tune | +0.142 |
commands | +0.142 |
Neg logit contributions | |
had | -0.143 |
missing | -0.142 |
made | -0.141 |
done | -0.137 |
wrong | -0.136 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
cess | +0.041 |
ck | +0.039 |
charged | +0.039 |
aret | +0.038 |
abbit | +0.038 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
missing | -0.012 |
sky | -0.011 |
made | -0.011 |
Token | a |
Token ID | 257 |
Feature activation | +1.69e+00 |
Pos logit contributions | |
wrong | +0.004 |
done | +0.004 |
made | +0.004 |
sky | +0.003 |
on | +0.003 |
Neg logit contributions | |
cess | -0.026 |
charged | -0.024 |
aret | -0.024 |
ck | -0.024 |
abbit | -0.023 |
Token | little |
Token ID | 1310 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
cess | +0.056 |
charged | +0.054 |
rums | +0.053 |
ck | +0.053 |
aret | +0.052 |
Neg logit contributions | |
wrong | -0.021 |
can | -0.021 |
had | -0.021 |
big | -0.020 |
sky | -0.020 |
Token | bird |
Token ID | 6512 |
Feature activation | +3.79e+00 |
Pos logit contributions | |
cess | +0.111 |
charged | +0.101 |
ck | +0.101 |
aret | +0.098 |
rums | +0.098 |
Neg logit contributions | |
sky | -0.065 |
can | -0.064 |
wrong | -0.063 |
made | -0.063 |
had | -0.061 |
Token | ie |
Token ID | 494 |
Feature activation | +3.75e+00 |
Pos logit contributions | |
cess | +0.112 |
ishment | +0.105 |
eter | +0.103 |
rily | +0.101 |
aret | +0.101 |
Neg logit contributions | |
made | -0.072 |
on | -0.071 |
had | -0.067 |
wrong | -0.066 |
all | -0.066 |
Token | named |
Token ID | 3706 |
Feature activation | +2.81e+00 |
Pos logit contributions | |
cess | +0.125 |
ishment | +0.120 |
eter | +0.119 |
rums | +0.115 |
aret | +0.113 |
Neg logit contributions | |
made | -0.056 |
on | -0.056 |
had | -0.055 |
wrong | -0.052 |
all | -0.051 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.27e+00 |
Pos logit contributions | |
cess | +0.081 |
ymes | +0.074 |
charged | +0.074 |
rums | +0.072 |
oing | +0.072 |
Neg logit contributions | |
all | -0.054 |
wrong | -0.053 |
sky | -0.052 |
made | -0.051 |
missing | -0.051 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.57e+00 |
Pos logit contributions | |
cess | +0.221 |
rums | +0.194 |
rily | +0.191 |
wrists | +0.190 |
tune | +0.184 |
Neg logit contributions | |
m | -0.148 |
missing | -0.146 |
had | -0.139 |
wrong | -0.134 |
made | -0.133 |
Token | . |
Token ID | 13 |
Feature activation | -4.35e-01 |
Pos logit contributions | |
cess | +0.132 |
ishment | +0.123 |
eter | +0.121 |
rily | +0.121 |
charged | +0.119 |
Neg logit contributions | |
all | -0.038 |
made | -0.038 |
wrong | -0.038 |
had | -0.037 |
on | -0.037 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.18e+00 |
Pos logit contributions | |
done | +0.014 |
wrong | +0.014 |
made | +0.013 |
missing | +0.013 |
sky | +0.013 |
Neg logit contributions | |
cess | -0.006 |
charged | -0.005 |
ck | -0.005 |
abbit | -0.004 |
aret | -0.004 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
cess | +0.173 |
wrists | +0.145 |
tune | +0.142 |
commands | +0.141 |
keys | +0.140 |
Neg logit contributions | |
had | -0.147 |
missing | -0.147 |
wrong | -0.144 |
made | -0.143 |
done | -0.142 |
Token | lived |
Token ID | 5615 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
cess | +0.046 |
charged | +0.043 |
ck | +0.043 |
aret | +0.042 |
abbit | +0.041 |
Neg logit contributions | |
wrong | -0.014 |
done | -0.014 |
missing | -0.013 |
sky | -0.013 |
made | -0.012 |
Token | back |
Token ID | 736 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
cess | +0.019 |
ck | +0.017 |
charged | +0.017 |
aret | +0.017 |
abbit | +0.016 |
Neg logit contributions | |
wrong | -0.008 |
done | -0.008 |
sky | -0.008 |
can | -0.008 |
made | -0.007 |
Token | home |
Token ID | 1363 |
Feature activation | +9.55e-01 |
Pos logit contributions | |
cess | +0.067 |
charged | +0.063 |
aret | +0.062 |
ck | +0.062 |
oxy | +0.061 |
Neg logit contributions | |
wrong | -0.025 |
done | -0.025 |
on | -0.025 |
made | -0.024 |
missing | -0.024 |
Token | . |
Token ID | 13 |
Feature activation | +6.89e-01 |
Pos logit contributions | |
cess | +0.030 |
charged | +0.028 |
ck | +0.028 |
oxy | +0.028 |
abbit | +0.028 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
missing | -0.013 |
made | -0.013 |
on | -0.012 |
Token | She |
Token ID | 1375 |
Feature activation | -5.11e-01 |
Pos logit contributions | |
cess | +0.020 |
charged | +0.018 |
aret | +0.018 |
abbit | +0.018 |
ck | +0.018 |
Neg logit contributions | |
done | -0.011 |
wrong | -0.011 |
made | -0.010 |
missing | -0.010 |
sky | -0.010 |
Token | thanked |
Token ID | 26280 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
wrong | +0.008 |
done | +0.008 |
sky | +0.007 |
missing | +0.007 |
made | +0.007 |
Neg logit contributions | |
cess | -0.015 |
charged | -0.014 |
ck | -0.014 |
aret | -0.014 |
abbit | -0.014 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.27e+00 |
Pos logit contributions | |
cess | +0.084 |
charged | +0.083 |
ymes | +0.082 |
abbit | +0.081 |
eter | +0.081 |
Neg logit contributions | |
all | -0.037 |
made | -0.033 |
wrong | -0.033 |
on | -0.033 |
done | -0.031 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.28e+00 |
Pos logit contributions | |
cess | +0.185 |
rums | +0.173 |
wrists | +0.171 |
keys | +0.165 |
ck | +0.163 |
Neg logit contributions | |
m | -0.161 |
made | -0.157 |
missing | -0.153 |
had | -0.151 |
best | -0.148 |
Token | for |
Token ID | 329 |
Feature activation | +1.63e+00 |
Pos logit contributions | |
cess | +0.133 |
charged | +0.129 |
oxy | +0.127 |
ymes | +0.126 |
rums | +0.125 |
Neg logit contributions | |
all | -0.019 |
on | -0.017 |
best | -0.014 |
wrong | -0.014 |
made | -0.013 |
Token | her |
Token ID | 607 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
cess | +0.042 |
ck | +0.037 |
charged | +0.037 |
aret | +0.036 |
abbit | +0.035 |
Neg logit contributions | |
wrong | -0.032 |
done | -0.031 |
sky | -0.031 |
can | -0.030 |
made | -0.030 |
Token | help |
Token ID | 1037 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
cess | +0.032 |
charged | +0.029 |
ck | +0.029 |
aret | +0.028 |
oxy | +0.028 |
Neg logit contributions | |
wrong | -0.021 |
done | -0.020 |
made | -0.020 |
best | -0.020 |
missing | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +3.55e+00 |
Pos logit contributions | |
cess | +0.055 |
charged | +0.053 |
ck | +0.051 |
ymes | +0.051 |
aret | +0.050 |
Neg logit contributions | |
made | -0.020 |
done | -0.019 |
wrong | -0.019 |
on | -0.019 |
all | -0.018 |
Token | Hello |
Token ID | 15496 |
Feature activation | +1.34e+00 |
Pos logit contributions | |
cess | +0.100 |
charged | +0.097 |
aret | +0.094 |
ck | +0.093 |
abbit | +0.092 |
Neg logit contributions | |
sky | -0.023 |
made | -0.022 |
had | -0.021 |
can | -0.017 |
wrong | -0.016 |
Token | ! |
Token ID | 0 |
Feature activation | +5.96e-01 |
Pos logit contributions | |
cess | +0.047 |
charged | +0.045 |
ck | +0.044 |
aret | +0.044 |
oxy | +0.043 |
Neg logit contributions | |
wrong | -0.014 |
can | -0.013 |
made | -0.013 |
sky | -0.013 |
done | -0.012 |
Token | My |
Token ID | 2011 |
Feature activation | +5.39e-01 |
Pos logit contributions | |
cess | +0.018 |
charged | +0.016 |
ck | +0.016 |
aret | +0.016 |
abbit | +0.016 |
Neg logit contributions | |
wrong | -0.009 |
done | -0.009 |
made | -0.009 |
sky | -0.009 |
can | -0.008 |
Token | name |
Token ID | 1438 |
Feature activation | -8.15e-01 |
Pos logit contributions | |
cess | +0.019 |
charged | +0.018 |
ck | +0.018 |
aret | +0.018 |
rily | +0.017 |
Neg logit contributions | |
best | -0.005 |
wrong | -0.005 |
sky | -0.005 |
can | -0.005 |
missing | -0.005 |
Token | is |
Token ID | 318 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
done | +0.006 |
wrong | +0.006 |
missing | +0.005 |
sky | +0.005 |
boards | +0.004 |
Neg logit contributions | |
cess | -0.030 |
ck | -0.027 |
charged | -0.027 |
ymes | -0.027 |
oxy | -0.027 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.23e+00 |
Pos logit contributions | |
cess | +0.013 |
charged | +0.012 |
ck | +0.012 |
aret | +0.012 |
ishment | +0.012 |
Neg logit contributions | |
wrong | -0.010 |
done | -0.010 |
made | -0.010 |
had | -0.009 |
all | -0.009 |
Token | ety |
Token ID | 2963 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
cess | +0.194 |
rily | +0.179 |
rums | +0.177 |
wrists | +0.176 |
aret | +0.174 |
Neg logit contributions | |
missing | -0.156 |
m | -0.155 |
had | -0.151 |
can | -0.150 |
made | -0.150 |
Token | . |
Token ID | 13 |
Feature activation | +8.66e-01 |
Pos logit contributions | |
cess | +0.130 |
ck | +0.123 |
charged | +0.121 |
ishment | +0.120 |
aret | +0.119 |
Neg logit contributions | |
wrong | -0.061 |
best | -0.056 |
on | -0.056 |
can | -0.055 |
made | -0.055 |
Token | What |
Token ID | 1867 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
cess | +0.023 |
charged | +0.020 |
abbit | +0.020 |
aret | +0.020 |
ck | +0.020 |
Neg logit contributions | |
done | -0.017 |
wrong | -0.016 |
made | -0.015 |
missing | -0.015 |
sky | -0.015 |
Token | are |
Token ID | 389 |
Feature activation | +1.65e-01 |
Pos logit contributions | |
wrong | +0.026 |
done | +0.026 |
made | +0.024 |
sky | +0.023 |
missing | +0.022 |
Neg logit contributions | |
cess | -0.118 |
charged | -0.110 |
ck | -0.109 |
aret | -0.108 |
abbit | -0.107 |
Token | you |
Token ID | 345 |
Feature activation | +3.31e-01 |
Pos logit contributions | |
cess | +0.004 |
charged | +0.004 |
ck | +0.004 |
aret | +0.004 |
abbit | +0.004 |
Neg logit contributions | |
wrong | -0.003 |
done | -0.003 |
made | -0.003 |
sky | -0.003 |
missing | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +4.00e-02 |
Pos logit contributions | |
cess | +0.020 |
charged | +0.018 |
aret | +0.018 |
abbit | +0.018 |
ck | +0.018 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
made | -0.012 |
missing | -0.012 |
sky | -0.012 |
Token | little |
Token ID | 1310 |
Feature activation | +2.03e+00 |
Pos logit contributions | |
cess | +0.001 |
charged | +0.001 |
ck | +0.001 |
aret | +0.001 |
abbit | +0.001 |
Neg logit contributions | |
wrong | -0.001 |
done | -0.001 |
made | -0.001 |
sky | -0.001 |
missing | -0.001 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.94e+00 |
Pos logit contributions | |
cess | +0.058 |
charged | +0.053 |
aret | +0.053 |
oxy | +0.052 |
ck | +0.052 |
Neg logit contributions | |
wrong | -0.037 |
made | -0.036 |
can | -0.035 |
on | -0.034 |
done | -0.034 |
Token | ie |
Token ID | 494 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
cess | +0.097 |
charged | +0.089 |
aret | +0.089 |
abbit | +0.087 |
ishment | +0.086 |
Neg logit contributions | |
wrong | -0.042 |
made | -0.041 |
on | -0.040 |
done | -0.039 |
can | -0.038 |
Token | like |
Token ID | 588 |
Feature activation | -2.65e-01 |
Pos logit contributions | |
cess | +0.068 |
charged | +0.063 |
ck | +0.062 |
aret | +0.061 |
abbit | +0.060 |
Neg logit contributions | |
wrong | -0.033 |
done | -0.032 |
made | -0.031 |
can | -0.031 |
had | -0.031 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.22e+00 |
Pos logit contributions | |
done | +0.004 |
wrong | +0.004 |
made | +0.004 |
sky | +0.004 |
missing | +0.004 |
Neg logit contributions | |
cess | -0.008 |
charged | -0.007 |
ck | -0.007 |
aret | -0.007 |
abbit | -0.007 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
cess | +0.217 |
rily | +0.199 |
aret | +0.194 |
God | +0.190 |
Amen | +0.189 |
Neg logit contributions | |
wrong | -0.143 |
missing | -0.143 |
m | -0.140 |
done | -0.137 |
made | -0.137 |
Token | . |
Token ID | 13 |
Feature activation | +7.85e-01 |
Pos logit contributions | |
cess | +0.099 |
charged | +0.089 |
aret | +0.089 |
ck | +0.087 |
abbit | +0.087 |
Neg logit contributions | |
wrong | -0.047 |
done | -0.045 |
made | -0.044 |
can | -0.044 |
had | -0.043 |
Token | So |
Token ID | 1406 |
Feature activation | -1.42e+00 |
Pos logit contributions | |
cess | +0.027 |
ck | +0.026 |
charged | +0.026 |
aret | +0.025 |
ymes | +0.025 |
Neg logit contributions | |
wrong | -0.008 |
sky | -0.008 |
can | -0.008 |
done | -0.008 |
had | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
done | +0.020 |
wrong | +0.020 |
sky | +0.019 |
boards | +0.019 |
missing | +0.019 |
Neg logit contributions | |
cess | -0.042 |
ck | -0.039 |
aret | -0.037 |
charged | -0.037 |
abbit | -0.037 |
Token | Twe |
Token ID | 24205 |
Feature activation | +7.26e+00 |
Pos logit contributions | |
done | +0.029 |
wrong | +0.029 |
sky | +0.028 |
made | +0.027 |
missing | +0.027 |
Neg logit contributions | |
cess | -0.027 |
ck | -0.024 |
charged | -0.024 |
aret | -0.023 |
oxy | -0.023 |
Token | how |
Token ID | 703 |
Feature activation | -1.63e+00 |
Pos logit contributions | |
cess | +0.030 |
chick | +0.027 |
swat | +0.026 |
ck | +0.026 |
God | +0.025 |
Neg logit contributions | |
done | -0.022 |
boards | -0.021 |
sky | -0.020 |
missing | -0.019 |
wrong | -0.018 |
Token | are |
Token ID | 389 |
Feature activation | +4.10e-01 |
Pos logit contributions | |
sky | +0.020 |
missing | +0.019 |
toilet | +0.017 |
boards | +0.017 |
bottom | +0.016 |
Neg logit contributions | |
ymes | -0.047 |
cess | -0.045 |
oing | -0.044 |
charged | -0.044 |
ck | -0.044 |
Token | you |
Token ID | 345 |
Feature activation | +1.91e+00 |
Pos logit contributions | |
cess | +0.010 |
charged | +0.009 |
ck | +0.009 |
aret | +0.009 |
abbit | +0.009 |
Neg logit contributions | |
wrong | -0.008 |
done | -0.008 |
made | -0.008 |
sky | -0.008 |
missing | -0.008 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
cess | +0.056 |
charged | +0.051 |
ck | +0.050 |
abbit | +0.048 |
feat | +0.048 |
Neg logit contributions | |
sky | -0.027 |
boards | -0.027 |
wrong | -0.026 |
has | -0.025 |
had | -0.024 |
Token | asked |
Token ID | 1965 |
Feature activation | +6.16e-01 |
Pos logit contributions | |
cess | +0.061 |
charged | +0.057 |
ymes | +0.056 |
eter | +0.056 |
aret | +0.055 |
Neg logit contributions | |
made | -0.036 |
had | -0.036 |
wrong | -0.036 |
all | -0.034 |
missing | -0.033 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.21e+00 |
Pos logit contributions | |
cess | +0.017 |
charged | +0.016 |
aret | +0.015 |
ymes | +0.015 |
oing | +0.015 |
Neg logit contributions | |
all | -0.012 |
on | -0.012 |
wrong | -0.012 |
made | -0.012 |
had | -0.011 |
Token | ety |
Token ID | 2963 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
cess | +0.194 |
wrists | +0.179 |
keys | +0.176 |
aret | +0.173 |
tunes | +0.169 |
Neg logit contributions | |
m | -0.168 |
| -0.146 |
can | -0.145 |
made | -0.144 |
missing | -0.143 |
Token | . |
Token ID | 13 |
Feature activation | +7.43e-01 |
Pos logit contributions | |
oxy | +0.157 |
cle | +0.157 |
cess | +0.155 |
ymes | +0.154 |
aret | +0.153 |
Neg logit contributions | |
all | -0.036 |
on | -0.036 |
wrong | -0.034 |
missing | -0.030 |
made | -0.029 |
Token |
|
Token ID | 198 |
Feature activation | +5.10e-01 |
Pos logit contributions | |
cess | +0.025 |
charged | +0.023 |
ck | +0.023 |
aret | +0.022 |
abbit | +0.022 |
Neg logit contributions | |
wrong | -0.009 |
done | -0.009 |
made | -0.008 |
sky | -0.008 |
missing | -0.008 |
Token |
|
Token ID | 198 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
cess | +0.019 |
charged | +0.018 |
ishment | +0.018 |
ck | +0.018 |
eter | +0.018 |
Neg logit contributions | |
wrong | -0.004 |
sky | -0.004 |
had | -0.004 |
can | -0.003 |
made | -0.003 |
Token | " |
Token ID | 1 |
Feature activation | +1.06e+00 |
Pos logit contributions | |
cess | +0.049 |
charged | +0.046 |
aret | +0.045 |
eter | +0.045 |
ishment | +0.044 |
Neg logit contributions | |
made | -0.031 |
had | -0.031 |
wrong | -0.031 |
can | -0.030 |
sky | -0.030 |
Token | save |
Token ID | 3613 |
Feature activation | +1.65e+00 |
Pos logit contributions | |
cess | +0.088 |
charged | +0.080 |
aret | +0.078 |
ck | +0.078 |
abbit | +0.078 |
Neg logit contributions | |
done | -0.058 |
wrong | -0.058 |
made | -0.056 |
missing | -0.054 |
can | -0.054 |
Token | the |
Token ID | 262 |
Feature activation | +2.79e-01 |
Pos logit contributions | |
cess | +0.051 |
charged | +0.046 |
aret | +0.045 |
ck | +0.045 |
abbit | +0.045 |
Neg logit contributions | |
wrong | -0.025 |
done | -0.024 |
made | -0.023 |
can | -0.022 |
sky | -0.022 |
Token | baby |
Token ID | 5156 |
Feature activation | +3.16e+00 |
Pos logit contributions | |
cess | +0.007 |
charged | +0.007 |
ck | +0.006 |
aret | +0.006 |
abbit | +0.006 |
Neg logit contributions | |
wrong | -0.006 |
done | -0.006 |
made | -0.005 |
sky | -0.005 |
missing | -0.005 |
Token | birds |
Token ID | 10087 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
cess | +0.080 |
charged | +0.072 |
aret | +0.071 |
ck | +0.070 |
abbit | +0.069 |
Neg logit contributions | |
wrong | -0.065 |
done | -0.064 |
made | -0.064 |
can | -0.063 |
sky | -0.062 |
Token | !" |
Token ID | 2474 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
cess | +0.075 |
charged | +0.068 |
ck | +0.067 |
aret | +0.066 |
abbit | +0.065 |
Neg logit contributions | |
wrong | -0.050 |
done | -0.050 |
made | -0.048 |
sky | -0.047 |
missing | -0.047 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.20e+00 |
Pos logit contributions | |
cess | +0.042 |
charged | +0.039 |
eter | +0.038 |
ymes | +0.038 |
aret | +0.038 |
Neg logit contributions | |
wrong | -0.028 |
made | -0.027 |
had | -0.027 |
done | -0.026 |
all | -0.026 |
Token | ety |
Token ID | 2963 |
Feature activation | +2.61e+00 |
Pos logit contributions | |
cess | +0.203 |
tunes | +0.201 |
tune | +0.196 |
films | +0.194 |
hospitality | +0.189 |
Neg logit contributions | |
m | -0.140 |
had | -0.126 |
| -0.122 |
missing | -0.118 |
wrong | -0.117 |
Token | said |
Token ID | 531 |
Feature activation | +1.03e-01 |
Pos logit contributions | |
ymes | +0.089 |
charged | +0.089 |
cess | +0.089 |
ishment | +0.088 |
eter | +0.085 |
Neg logit contributions | |
had | -0.032 |
made | -0.032 |
done | -0.027 |
was | -0.027 |
two | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
charged | +0.004 |
ymes | +0.004 |
cess | +0.004 |
ishment | +0.004 |
oxy | +0.003 |
Neg logit contributions | |
on | -0.001 |
all | -0.001 |
wrong | -0.001 |
right | -0.001 |
made | -0.001 |
Token | |
Token ID | 220 |
Feature activation | -1.84e+00 |
Pos logit contributions | |
cess | +0.029 |
charged | +0.027 |
ck | +0.027 |
ymes | +0.027 |
aret | +0.026 |
Neg logit contributions | |
wrong | -0.007 |
sky | -0.007 |
done | -0.006 |
had | -0.006 |
made | -0.006 |
Token |
|
Token ID | 198 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
wrong | +0.019 |
done | +0.018 |
sky | +0.016 |
made | +0.016 |
missing | +0.016 |
Neg logit contributions | |
cess | -0.065 |
charged | -0.061 |
ck | -0.060 |
aret | -0.059 |
abbit | -0.058 |
Token | , |
Token ID | 11 |
Feature activation | -4.80e-01 |
Pos logit contributions | |
cess | +0.010 |
charged | +0.009 |
ck | +0.009 |
aret | +0.009 |
ishment | +0.009 |
Neg logit contributions | |
made | -0.003 |
done | -0.003 |
wrong | -0.003 |
had | -0.003 |
can | -0.003 |
Token | a |
Token ID | 257 |
Feature activation | +7.66e-02 |
Pos logit contributions | |
wrong | +0.005 |
made | +0.005 |
done | +0.005 |
had | +0.005 |
on | +0.005 |
Neg logit contributions | |
cess | -0.017 |
charged | -0.016 |
ck | -0.015 |
aret | -0.015 |
eter | -0.015 |
Token | little |
Token ID | 1310 |
Feature activation | +2.21e+00 |
Pos logit contributions | |
cess | +0.002 |
charged | +0.002 |
abbit | +0.002 |
aret | +0.002 |
ck | +0.002 |
Neg logit contributions | |
done | -0.001 |
wrong | -0.001 |
made | -0.001 |
sky | -0.001 |
missing | -0.001 |
Token | bird |
Token ID | 6512 |
Feature activation | +2.43e+00 |
Pos logit contributions | |
cess | +0.057 |
ck | +0.051 |
charged | +0.051 |
aret | +0.050 |
oxy | +0.049 |
Neg logit contributions | |
wrong | -0.046 |
made | -0.046 |
can | -0.045 |
sky | -0.045 |
on | -0.044 |
Token | named |
Token ID | 3706 |
Feature activation | +1.49e+00 |
Pos logit contributions | |
cess | +0.050 |
charged | +0.044 |
aret | +0.042 |
abbit | +0.041 |
oxy | +0.041 |
Neg logit contributions | |
wrong | -0.061 |
done | -0.061 |
made | -0.061 |
had | -0.059 |
can | -0.059 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.19e+00 |
Pos logit contributions | |
cess | +0.037 |
charged | +0.037 |
ck | +0.037 |
oing | +0.035 |
tune | +0.035 |
Neg logit contributions | |
wrong | -0.029 |
made | -0.028 |
can | -0.028 |
had | -0.027 |
all | -0.027 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
cess | +0.201 |
wrists | +0.181 |
rily | +0.179 |
rums | +0.170 |
tune | +0.168 |
Neg logit contributions | |
m | -0.158 |
missing | -0.156 |
had | -0.152 |
can | -0.151 |
made | -0.151 |
Token | , |
Token ID | 11 |
Feature activation | -3.80e-01 |
Pos logit contributions | |
cess | +0.064 |
charged | +0.059 |
ck | +0.058 |
aret | +0.058 |
abbit | +0.057 |
Neg logit contributions | |
wrong | -0.020 |
done | -0.019 |
made | -0.018 |
sky | -0.018 |
missing | -0.017 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
wrong | +0.005 |
done | +0.004 |
sky | +0.004 |
best | +0.004 |
missing | +0.004 |
Neg logit contributions | |
cess | -0.012 |
ck | -0.011 |
charged | -0.011 |
abbit | -0.011 |
chick | -0.011 |
Token | him |
Token ID | 683 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
done | +0.029 |
sky | +0.027 |
wrong | +0.027 |
missing | +0.026 |
made | +0.026 |
Neg logit contributions | |
cess | -0.035 |
charged | -0.031 |
ck | -0.031 |
aret | -0.030 |
abbit | -0.030 |
Token | struggling |
Token ID | 9648 |
Feature activation | +8.10e-03 |
Pos logit contributions | |
cess | +0.033 |
charged | +0.030 |
aret | +0.030 |
ck | +0.030 |
oxy | +0.029 |
Neg logit contributions | |
wrong | -0.017 |
done | -0.017 |
made | -0.016 |
missing | -0.016 |
had | -0.016 |
Token | there |
Token ID | 612 |
Feature activation | -4.61e+00 |
Pos logit contributions | |
done | +0.020 |
boards | +0.020 |
sky | +0.020 |
wrong | +0.019 |
missing | +0.018 |
Neg logit contributions | |
cess | -0.061 |
ck | -0.060 |
abbit | -0.058 |
charged | -0.057 |
chick | -0.057 |
Token | was |
Token ID | 373 |
Feature activation | -6.84e-01 |
Pos logit contributions | |
wrong | +0.053 |
done | +0.049 |
missing | +0.048 |
sky | +0.046 |
boards | +0.042 |
Neg logit contributions | |
cess | -0.145 |
ck | -0.136 |
charged | -0.131 |
oxy | -0.130 |
abbit | -0.130 |
Token | a |
Token ID | 257 |
Feature activation | +1.89e+00 |
Pos logit contributions | |
wrong | +0.004 |
done | +0.004 |
made | +0.004 |
sky | +0.003 |
on | +0.003 |
Neg logit contributions | |
cess | -0.028 |
charged | -0.026 |
aret | -0.026 |
ck | -0.026 |
abbit | -0.025 |
Token | bird |
Token ID | 6512 |
Feature activation | +4.26e+00 |
Pos logit contributions | |
cess | +0.064 |
charged | +0.062 |
rums | +0.061 |
ck | +0.060 |
aret | +0.060 |
Neg logit contributions | |
wrong | -0.022 |
big | -0.022 |
can | -0.022 |
had | -0.022 |
made | -0.022 |
Token | named |
Token ID | 3706 |
Feature activation | +2.79e+00 |
Pos logit contributions | |
cess | +0.128 |
ishment | +0.119 |
eter | +0.117 |
charged | +0.115 |
aret | +0.115 |
Neg logit contributions | |
made | -0.080 |
on | -0.079 |
all | -0.074 |
had | -0.073 |
wrong | -0.073 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
cess | +0.081 |
charged | +0.080 |
ymes | +0.078 |
rums | +0.075 |
aret | +0.075 |
Neg logit contributions | |
all | -0.049 |
wrong | -0.048 |
made | -0.047 |
sky | -0.047 |
big | -0.046 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.68e+00 |
Pos logit contributions | |
cess | +0.208 |
rums | +0.186 |
rily | +0.180 |
wrists | +0.178 |
tune | +0.176 |
Neg logit contributions | |
missing | -0.158 |
had | -0.153 |
m | -0.151 |
made | -0.148 |
wrong | -0.148 |
Token | . |
Token ID | 13 |
Feature activation | -3.56e-01 |
Pos logit contributions | |
cess | +0.138 |
ishment | +0.132 |
eter | +0.129 |
rily | +0.127 |
charged | +0.127 |
Neg logit contributions | |
made | -0.039 |
had | -0.038 |
all | -0.038 |
on | -0.037 |
wrong | -0.035 |
Token | Twe |
Token ID | 24205 |
Feature activation | +6.77e+00 |
Pos logit contributions | |
done | +0.012 |
wrong | +0.012 |
made | +0.011 |
missing | +0.011 |
can | +0.011 |
Neg logit contributions | |
cess | -0.004 |
charged | -0.003 |
ck | -0.003 |
aret | -0.003 |
abbit | -0.003 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
cess | +0.150 |
wrists | +0.125 |
aret | +0.125 |
tune | +0.122 |
charged | +0.122 |
Neg logit contributions | |
had | -0.155 |
missing | -0.153 |
made | -0.152 |
wrong | -0.152 |
done | -0.151 |
Token | lived |
Token ID | 5615 |
Feature activation | -1.08e+00 |
Pos logit contributions | |
cess | +0.046 |
ck | +0.043 |
charged | +0.043 |
aret | +0.042 |
abbit | +0.041 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
missing | -0.012 |
sky | -0.012 |
made | -0.011 |
Token | to |
Token ID | 284 |
Feature activation | +4.24e+00 |
Pos logit contributions | |
cess | +0.026 |
ck | +0.023 |
charged | +0.023 |
aret | +0.022 |
oxy | +0.022 |
Neg logit contributions | |
done | -0.013 |
sky | -0.012 |
wrong | -0.012 |
can | -0.012 |
made | -0.012 |
Token | never |
Token ID | 1239 |
Feature activation | +1.61e+00 |
Pos logit contributions | |
cess | +0.121 |
charged | +0.110 |
ck | +0.109 |
aret | +0.107 |
abbit | +0.105 |
Neg logit contributions | |
wrong | -0.071 |
done | -0.070 |
sky | -0.067 |
missing | -0.067 |
made | -0.067 |
Token | try |
Token ID | 1949 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
cess | +0.045 |
charged | +0.041 |
ck | +0.041 |
aret | +0.040 |
abbit | +0.040 |
Neg logit contributions | |
wrong | -0.028 |
done | -0.028 |
made | -0.026 |
sky | -0.026 |
missing | -0.026 |
Token | to |
Token ID | 284 |
Feature activation | +2.47e+00 |
Pos logit contributions | |
done | +0.017 |
sky | +0.016 |
can | +0.016 |
wrong | +0.015 |
made | +0.015 |
Neg logit contributions | |
cess | -0.029 |
ck | -0.026 |
charged | -0.026 |
abbit | -0.025 |
oxy | -0.025 |
Token | catch |
Token ID | 4929 |
Feature activation | +1.05e+00 |
Pos logit contributions | |
cess | +0.070 |
aret | +0.064 |
abbit | +0.064 |
charged | +0.064 |
rily | +0.064 |
Neg logit contributions | |
made | -0.042 |
wrong | -0.042 |
done | -0.042 |
can | -0.040 |
had | -0.040 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.18e+00 |
Pos logit contributions | |
cess | +0.018 |
ck | +0.016 |
charged | +0.015 |
abbit | +0.014 |
oxy | +0.014 |
Neg logit contributions | |
done | -0.030 |
wrong | -0.029 |
sky | -0.028 |
made | -0.028 |
missing | -0.028 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
cess | +0.183 |
rily | +0.181 |
aret | +0.171 |
wrists | +0.168 |
tune | +0.159 |
Neg logit contributions | |
m | -0.163 |
wrong | -0.161 |
made | -0.159 |
missing | -0.159 |
done | -0.158 |
Token | again |
Token ID | 757 |
Feature activation | +3.57e-01 |
Pos logit contributions | |
cess | +0.062 |
aret | +0.059 |
charged | +0.057 |
abbit | +0.057 |
cle | +0.056 |
Neg logit contributions | |
wrong | -0.029 |
on | -0.029 |
all | -0.026 |
missing | -0.026 |
made | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | +5.36e-01 |
Pos logit contributions | |
cess | +0.011 |
charged | +0.011 |
aret | +0.010 |
ck | +0.010 |
abbit | +0.010 |
Neg logit contributions | |
wrong | -0.005 |
done | -0.005 |
made | -0.005 |
on | -0.005 |
had | -0.005 |
Token | From |
Token ID | 3574 |
Feature activation | +3.83e-01 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.012 |
ck | +0.012 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
wrong | -0.011 |
done | -0.010 |
made | -0.010 |
sky | -0.010 |
can | -0.010 |
Token | that |
Token ID | 326 |
Feature activation | -2.96e+00 |
Pos logit contributions | |
cess | +0.014 |
ck | +0.013 |
charged | +0.013 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
done | -0.004 |
wrong | -0.003 |
sky | -0.003 |
missing | -0.003 |
made | -0.003 |
Token | my |
Token ID | 1820 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
cess | +0.069 |
charged | +0.062 |
ck | +0.062 |
aret | +0.062 |
abbit | +0.061 |
Neg logit contributions | |
wrong | -0.042 |
done | -0.041 |
sky | -0.039 |
missing | -0.039 |
made | -0.038 |
Token | ." |
Token ID | 526 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
cess | +0.042 |
charged | +0.038 |
ck | +0.038 |
aret | +0.038 |
abbit | +0.037 |
Neg logit contributions | |
wrong | -0.023 |
done | -0.023 |
made | -0.022 |
sky | -0.021 |
missing | -0.021 |
Token | |
Token ID | 220 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
ymes | +0.093 |
ck | +0.093 |
charged | +0.093 |
cess | +0.093 |
ishment | +0.088 |
Neg logit contributions | |
had | -0.018 |
| -0.015 |
sky | -0.014 |
all | -0.014 |
made | -0.013 |
Token |
|
Token ID | 198 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.010 |
sky | +0.009 |
had | +0.009 |
can | +0.009 |
Neg logit contributions | |
cess | -0.046 |
charged | -0.043 |
ck | -0.043 |
tune | -0.042 |
ishment | -0.042 |
Token |
|
Token ID | 198 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
cess | +0.080 |
ishment | +0.079 |
charged | +0.079 |
eter | +0.077 |
chicks | +0.076 |
Neg logit contributions | |
| -0.007 |
had | -0.005 |
was | -0.003 |
sky | -0.003 |
two | -0.003 |
Token | Twe |
Token ID | 32665 |
Feature activation | +8.15e+00 |
Pos logit contributions | |
cess | +0.066 |
ishment | +0.065 |
charged | +0.064 |
rily | +0.063 |
eter | +0.062 |
Neg logit contributions | |
had | -0.042 |
It | -0.038 |
two | -0.038 |
| -0.036 |
made | -0.036 |
Token | ety |
Token ID | 2963 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
cess | +0.174 |
chicks | +0.156 |
keys | +0.156 |
aret | +0.155 |
rums | +0.155 |
Neg logit contributions | |
had | -0.170 |
missing | -0.168 |
done | -0.163 |
wrong | -0.162 |
m | -0.161 |
Token | flew |
Token ID | 13112 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
cess | +0.052 |
charged | +0.048 |
ck | +0.047 |
aret | +0.047 |
abbit | +0.047 |
Neg logit contributions | |
done | -0.021 |
made | -0.020 |
wrong | -0.020 |
had | -0.020 |
can | -0.019 |
Token | around |
Token ID | 1088 |
Feature activation | +3.53e-01 |
Pos logit contributions | |
done | +0.028 |
wrong | +0.028 |
can | +0.026 |
has | +0.026 |
sky | +0.026 |
Neg logit contributions | |
cess | -0.050 |
ck | -0.046 |
aret | -0.046 |
rily | -0.046 |
charged | -0.045 |
Token | the |
Token ID | 262 |
Feature activation | -1.85e+00 |
Pos logit contributions | |
cess | +0.011 |
charged | +0.010 |
ck | +0.010 |
aret | +0.010 |
oxy | +0.010 |
Neg logit contributions | |
wrong | -0.005 |
done | -0.005 |
made | -0.005 |
sky | -0.005 |
missing | -0.005 |
Token | forest |
Token ID | 8222 |
Feature activation | +5.82e-01 |
Pos logit contributions | |
done | +0.031 |
wrong | +0.027 |
made | +0.027 |
had | +0.026 |
best | +0.025 |
Neg logit contributions | |
cess | -0.054 |
charged | -0.049 |
chick | -0.048 |
tune | -0.048 |
abbit | -0.048 |
Token | a |
Token ID | 257 |
Feature activation | -2.26e-01 |
Pos logit contributions | |
done | +0.045 |
wrong | +0.045 |
can | +0.045 |
boards | +0.044 |
sky | +0.040 |
Neg logit contributions | |
cess | -0.117 |
oxy | -0.112 |
ck | -0.108 |
abbit | -0.108 |
charged | -0.107 |
Token | film |
Token ID | 2646 |
Feature activation | -5.79e-01 |
Pos logit contributions | |
done | +0.003 |
wrong | +0.003 |
sky | +0.003 |
boards | +0.003 |
made | +0.003 |
Neg logit contributions | |
cess | -0.007 |
abbit | -0.006 |
chick | -0.006 |
aret | -0.006 |
charged | -0.006 |
Token | camera |
Token ID | 4676 |
Feature activation | -4.46e-01 |
Pos logit contributions | |
done | +0.007 |
wrong | +0.007 |
sky | +0.007 |
can | +0.007 |
missing | +0.006 |
Neg logit contributions | |
cess | -0.017 |
ck | -0.016 |
abbit | -0.016 |
tune | -0.016 |
chick | -0.016 |
Token | and |
Token ID | 290 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
done | +0.006 |
wrong | +0.005 |
can | +0.005 |
am | +0.005 |
has | +0.005 |
Neg logit contributions | |
cess | -0.014 |
abbit | -0.012 |
ck | -0.012 |
tune | -0.012 |
oxy | -0.012 |
Token | asked |
Token ID | 1965 |
Feature activation | -2.09e-02 |
Pos logit contributions | |
cess | +0.049 |
ck | +0.043 |
chick | +0.042 |
oxy | +0.042 |
aret | +0.042 |
Neg logit contributions | |
done | -0.030 |
wrong | -0.030 |
sky | -0.028 |
missing | -0.028 |
best | -0.026 |
Token | Twe |
Token ID | 24205 |
Feature activation | +8.14e+00 |
Pos logit contributions | |
wrong | +0.000 |
done | +0.000 |
made | +0.000 |
sky | +0.000 |
missing | +0.000 |
Neg logit contributions | |
cess | -0.001 |
charged | -0.001 |
ck | -0.001 |
aret | -0.001 |
abbit | -0.001 |
Token | ety |
Token ID | 2963 |
Feature activation | +3.53e+00 |
Pos logit contributions | |
cess | +0.200 |
rily | +0.176 |
aret | +0.174 |
wrists | +0.170 |
rums | +0.170 |
Neg logit contributions | |
m | -0.157 |
missing | -0.157 |
wrong | -0.157 |
made | -0.154 |
had | -0.153 |
Token | if |
Token ID | 611 |
Feature activation | -1.02e+00 |
Pos logit contributions | |
cess | +0.114 |
charged | +0.107 |
ck | +0.106 |
oxy | +0.105 |
aret | +0.104 |
Neg logit contributions | |
wrong | -0.047 |
done | -0.043 |
can | -0.043 |
made | -0.042 |
missing | -0.042 |
Token | she |
Token ID | 673 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
done | +0.016 |
sky | +0.015 |
wrong | +0.015 |
missing | +0.014 |
made | +0.014 |
Neg logit contributions | |
cess | -0.030 |
ck | -0.027 |
charged | -0.026 |
aret | -0.026 |
oxy | -0.026 |
Token | could |
Token ID | 714 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
made | +0.027 |
had | +0.027 |
can | +0.026 |
done | +0.024 |
wrong | +0.024 |
Neg logit contributions | |
cess | -0.138 |
charged | -0.131 |
aret | -0.129 |
ck | -0.126 |
abbit | -0.125 |
Token | take |
Token ID | 1011 |
Feature activation | -2.13e-01 |
Pos logit contributions | |
cess | +0.016 |
charged | +0.015 |
aret | +0.014 |
ck | +0.014 |
abbit | +0.014 |
Neg logit contributions | |
done | -0.010 |
made | -0.010 |
wrong | -0.010 |
can | -0.010 |
had | -0.010 |
Token | had |
Token ID | 550 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
cess | +0.068 |
ck | +0.063 |
tune | +0.061 |
charged | +0.060 |
aret | +0.059 |
Neg logit contributions | |
wrong | -0.044 |
sky | -0.041 |
| -0.040 |
done | -0.040 |
missing | -0.040 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
wrong | +0.026 |
sky | +0.025 |
done | +0.024 |
missing | +0.024 |
can | +0.023 |
Neg logit contributions | |
cess | -0.086 |
charged | -0.080 |
ck | -0.080 |
oxy | -0.079 |
eter | -0.079 |
Token | . |
Token ID | 13 |
Feature activation | -3.83e-01 |
Pos logit contributions | |
cess | +0.044 |
charged | +0.040 |
ck | +0.040 |
aret | +0.038 |
abbit | +0.038 |
Neg logit contributions | |
wrong | -0.030 |
done | -0.029 |
sky | -0.028 |
missing | -0.028 |
made | -0.028 |
Token | But |
Token ID | 887 |
Feature activation | +2.21e-01 |
Pos logit contributions | |
done | +0.008 |
wrong | +0.008 |
made | +0.007 |
missing | +0.007 |
sky | +0.007 |
Neg logit contributions | |
cess | -0.010 |
aret | -0.008 |
charged | -0.008 |
eter | -0.008 |
abbit | -0.008 |
Token | when |
Token ID | 618 |
Feature activation | -2.25e+00 |
Pos logit contributions | |
cess | +0.005 |
ck | +0.005 |
aret | +0.004 |
charged | +0.004 |
ishment | +0.004 |
Neg logit contributions | |
sky | -0.005 |
boards | -0.005 |
done | -0.005 |
wrong | -0.004 |
can | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | -7.70e+00 |
Pos logit contributions | |
boards | +0.045 |
sky | +0.043 |
done | +0.043 |
wrong | +0.042 |
can | +0.040 |
Neg logit contributions | |
cess | -0.051 |
ck | -0.043 |
aret | -0.043 |
charged | -0.042 |
tune | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -3.31e+00 |
Pos logit contributions | |
wrong | +0.075 |
sky | +0.070 |
best | +0.070 |
biggest | +0.069 |
can | +0.064 |
Neg logit contributions | |
cess | -0.246 |
ck | -0.235 |
tune | -0.221 |
charged | -0.215 |
aunted | -0.214 |
Token | time |
Token ID | 640 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
am | +0.043 |
can | +0.043 |
boards | +0.040 |
has | +0.040 |
sky | +0.037 |
Neg logit contributions | |
cess | -0.091 |
aret | -0.089 |
ck | -0.089 |
tune | -0.086 |
charged | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
sky | +0.081 |
biggest | +0.079 |
coolest | +0.079 |
woods | +0.076 |
toilet | +0.075 |
Neg logit contributions | |
aret | -0.061 |
cess | -0.060 |
fully | -0.055 |
rily | -0.055 |
ck | -0.053 |
Token | play |
Token ID | 711 |
Feature activation | -1.57e+00 |
Pos logit contributions | |
cess | +0.054 |
charged | +0.049 |
ck | +0.049 |
aret | +0.048 |
abbit | +0.047 |
Neg logit contributions | |
wrong | -0.028 |
done | -0.028 |
made | -0.027 |
sky | -0.026 |
missing | -0.026 |
Token | with |
Token ID | 351 |
Feature activation | +9.19e-01 |
Pos logit contributions | |
sky | +0.026 |
done | +0.025 |
has | +0.024 |
had | +0.023 |
made | +0.023 |
Neg logit contributions | |
cess | -0.043 |
tune | -0.037 |
charged | -0.036 |
rily | -0.035 |
aret | -0.035 |
Token | tw |
Token ID | 665 |
Feature activation | +1.33e+00 |
Pos logit contributions | |
cess | +0.014 |
ck | +0.013 |
charged | +0.013 |
wrists | +0.012 |
aret | +0.012 |
Neg logit contributions | |
wrong | -0.013 |
sky | -0.011 |
missing | -0.011 |
done | -0.011 |
best | -0.010 |
Token | irled |
Token ID | 49376 |
Feature activation | -6.47e-01 |
Pos logit contributions | |
cess | +0.029 |
charged | +0.025 |
abbit | +0.025 |
oxy | +0.025 |
aret | +0.024 |
Neg logit contributions | |
done | -0.032 |
made | -0.031 |
wrong | -0.030 |
on | -0.030 |
had | -0.029 |
Token | , |
Token ID | 11 |
Feature activation | +8.62e-01 |
Pos logit contributions | |
done | +0.011 |
wrong | +0.011 |
missing | +0.010 |
has | +0.010 |
am | +0.010 |
Neg logit contributions | |
cess | -0.017 |
tune | -0.016 |
ck | -0.016 |
rily | -0.016 |
aret | -0.015 |
Token | and |
Token ID | 290 |
Feature activation | +7.51e-01 |
Pos logit contributions | |
cess | +0.023 |
ck | +0.020 |
tune | +0.020 |
oing | +0.019 |
feat | +0.019 |
Neg logit contributions | |
wrong | -0.016 |
done | -0.015 |
am | -0.015 |
can | -0.015 |
| -0.014 |
Token | when |
Token ID | 618 |
Feature activation | -2.66e+00 |
Pos logit contributions | |
cess | +0.019 |
ck | +0.018 |
tune | +0.017 |
feat | +0.017 |
aret | +0.017 |
Neg logit contributions | |
wrong | -0.014 |
missing | -0.013 |
sky | -0.012 |
done | -0.012 |
| -0.012 |
Token | it |
Token ID | 340 |
Feature activation | -7.69e+00 |
Pos logit contributions | |
wrong | +0.042 |
done | +0.041 |
missing | +0.039 |
made | +0.038 |
can | +0.038 |
Neg logit contributions | |
cess | -0.074 |
charged | -0.067 |
ck | -0.066 |
aret | -0.065 |
oxy | -0.064 |
Token | was |
Token ID | 373 |
Feature activation | -3.71e+00 |
Pos logit contributions | |
wrong | +0.101 |
missing | +0.088 |
done | +0.080 |
living | +0.079 |
biggest | +0.079 |
Neg logit contributions | |
ck | -0.217 |
cess | -0.217 |
tune | -0.202 |
feat | -0.196 |
hr | -0.189 |
Token | time |
Token ID | 640 |
Feature activation | -4.22e+00 |
Pos logit contributions | |
am | +0.053 |
boards | +0.051 |
can | +0.051 |
has | +0.049 |
wrong | +0.046 |
Neg logit contributions | |
ck | -0.100 |
cess | -0.098 |
aret | -0.095 |
tune | -0.092 |
rily | -0.091 |
Token | to |
Token ID | 284 |
Feature activation | +1.31e+00 |
Pos logit contributions | |
am | +0.084 |
coolest | +0.083 |
toilet | +0.082 |
sky | +0.081 |
woods | +0.079 |
Neg logit contributions | |
aret | -0.076 |
rily | -0.075 |
cess | -0.072 |
fully | -0.070 |
heads | -0.068 |
Token | say |
Token ID | 910 |
Feature activation | -2.60e+00 |
Pos logit contributions | |
cess | +0.038 |
ck | +0.035 |
charged | +0.034 |
aret | +0.033 |
oxy | +0.032 |
Neg logit contributions | |
wrong | -0.022 |
done | -0.021 |
missing | -0.020 |
sky | -0.020 |
made | -0.020 |
Token | goodbye |
Token ID | 24829 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
missing | +0.056 |
done | +0.055 |
sky | +0.055 |
made | +0.053 |
had | +0.053 |
Neg logit contributions | |
cess | -0.058 |
tune | -0.047 |
aret | -0.046 |
ck | -0.045 |
.'" | -0.045 |
Token | her |
Token ID | 607 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
wrong | +0.001 |
done | +0.001 |
sky | +0.001 |
made | +0.001 |
missing | +0.001 |
Neg logit contributions | |
cess | -0.003 |
ck | -0.002 |
charged | -0.002 |
aret | -0.002 |
abbit | -0.002 |
Token | sauce |
Token ID | 10746 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
cess | +0.045 |
charged | +0.041 |
aret | +0.040 |
ck | +0.040 |
abbit | +0.039 |
Neg logit contributions | |
done | -0.028 |
wrong | -0.028 |
made | -0.027 |
missing | -0.026 |
can | -0.026 |
Token | , |
Token ID | 11 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
wrong | +0.034 |
sky | +0.034 |
done | +0.033 |
missing | +0.032 |
had | +0.032 |
Neg logit contributions | |
cess | -0.050 |
rily | -0.047 |
ck | -0.046 |
eter | -0.046 |
charged | -0.044 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
cess | +0.058 |
ck | +0.053 |
charged | +0.051 |
aret | +0.051 |
tune | +0.050 |
Neg logit contributions | |
wrong | -0.037 |
done | -0.037 |
sky | -0.035 |
can | -0.034 |
missing | -0.033 |
Token | when |
Token ID | 618 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
cess | +0.061 |
ck | +0.057 |
aret | +0.054 |
charged | +0.054 |
oxy | +0.054 |
Neg logit contributions | |
wrong | -0.038 |
sky | -0.036 |
done | -0.036 |
missing | -0.034 |
can | -0.034 |
Token | it |
Token ID | 340 |
Feature activation | -7.62e+00 |
Pos logit contributions | |
sky | +0.032 |
done | +0.032 |
wrong | +0.031 |
boards | +0.031 |
can | +0.030 |
Neg logit contributions | |
cess | -0.049 |
charged | -0.044 |
ck | -0.044 |
aret | -0.043 |
feat | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
wrong | +0.084 |
sky | +0.083 |
best | +0.077 |
biggest | +0.076 |
living | +0.072 |
Neg logit contributions | |
cess | -0.219 |
ck | -0.215 |
chick | -0.203 |
eter | -0.199 |
rily | -0.196 |
Token | done |
Token ID | 1760 |
Feature activation | -2.63e+00 |
Pos logit contributions | |
am | +0.035 |
sky | +0.035 |
can | +0.032 |
has | +0.032 |
boards | +0.031 |
Neg logit contributions | |
ck | -0.105 |
rily | -0.100 |
cess | -0.097 |
aret | -0.097 |
charged | -0.095 |
Token | , |
Token ID | 11 |
Feature activation | +6.12e-02 |
Pos logit contributions | |
sky | +0.046 |
has | +0.041 |
can | +0.041 |
done | +0.041 |
wrong | +0.041 |
Neg logit contributions | |
cess | -0.065 |
rily | -0.059 |
ck | -0.056 |
eter | -0.056 |
tune | -0.055 |
Token | she |
Token ID | 673 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
cess | +0.001 |
chick | +0.001 |
ck | +0.001 |
chicks | +0.001 |
rily | +0.001 |
Neg logit contributions | |
done | -0.001 |
wrong | -0.001 |
has | -0.001 |
sky | -0.001 |
can | -0.001 |
Token | put |
Token ID | 1234 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
wrong | +0.023 |
sky | +0.021 |
done | +0.021 |
best | +0.020 |
missing | +0.020 |
Neg logit contributions | |
cess | -0.052 |
ck | -0.049 |
oxy | -0.047 |
charged | -0.047 |
abbit | -0.046 |
Token | the |
Token ID | 262 |
Feature activation | -4.23e-01 |
Pos logit contributions | |
wrong | +0.007 |
done | +0.007 |
made | +0.007 |
sky | +0.007 |
missing | +0.007 |
Neg logit contributions | |
cess | -0.012 |
charged | -0.011 |
ck | -0.011 |
aret | -0.011 |
abbit | -0.011 |
Token | message |
Token ID | 3275 |
Feature activation | -2.59e+00 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.010 |
made | +0.010 |
sky | +0.009 |
can | +0.009 |
Neg logit contributions | |
cess | -0.009 |
charged | -0.008 |
ck | -0.008 |
aret | -0.008 |
abbit | -0.008 |
Token | , |
Token ID | 11 |
Feature activation | +3.79e-01 |
Pos logit contributions | |
done | +0.040 |
wrong | +0.040 |
sky | +0.039 |
missing | +0.038 |
| +0.037 |
Neg logit contributions | |
cess | -0.069 |
ck | -0.064 |
aret | -0.062 |
tune | -0.062 |
rily | -0.061 |
Token | but |
Token ID | 475 |
Feature activation | -9.79e-01 |
Pos logit contributions | |
cess | +0.010 |
ck | +0.009 |
charged | +0.009 |
aret | +0.009 |
abbit | +0.009 |
Neg logit contributions | |
wrong | -0.007 |
done | -0.007 |
sky | -0.006 |
missing | -0.006 |
best | -0.006 |
Token | when |
Token ID | 618 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
wrong | +0.016 |
done | +0.016 |
sky | +0.016 |
missing | +0.015 |
has | +0.015 |
Neg logit contributions | |
cess | -0.026 |
ck | -0.023 |
charged | -0.023 |
aret | -0.023 |
eter | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | -7.55e+00 |
Pos logit contributions | |
wrong | +0.037 |
done | +0.036 |
sky | +0.034 |
missing | +0.034 |
made | +0.034 |
Neg logit contributions | |
cess | -0.064 |
charged | -0.056 |
aret | -0.055 |
ck | -0.055 |
oxy | -0.055 |
Token | was |
Token ID | 373 |
Feature activation | -3.63e+00 |
Pos logit contributions | |
wrong | +0.082 |
sky | +0.079 |
best | +0.073 |
done | +0.072 |
biggest | +0.071 |
Neg logit contributions | |
cess | -0.229 |
ck | -0.225 |
chick | -0.205 |
eter | -0.204 |
charged | -0.203 |
Token | Molly |
Token ID | 30236 |
Feature activation | +8.59e-03 |
Pos logit contributions | |
am | +0.048 |
has | +0.047 |
can | +0.047 |
sky | +0.046 |
| +0.043 |
Neg logit contributions | |
cess | -0.102 |
ck | -0.101 |
aret | -0.099 |
ishment | -0.095 |
charged | -0.094 |
Token | 's |
Token ID | 338 |
Feature activation | -2.28e+00 |
Pos logit contributions | |
cess | +0.000 |
ck | +0.000 |
charged | +0.000 |
oxy | +0.000 |
abbit | +0.000 |
Neg logit contributions | |
sky | -0.000 |
wrong | -0.000 |
done | -0.000 |
best | -0.000 |
missing | -0.000 |
Token | turn |
Token ID | 1210 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
wrong | +0.045 |
done | +0.045 |
made | +0.043 |
missing | +0.043 |
sky | +0.043 |
Neg logit contributions | |
cess | -0.060 |
charged | -0.054 |
ck | -0.053 |
aret | -0.052 |
oxy | -0.051 |
Token | , |
Token ID | 11 |
Feature activation | -6.44e-01 |
Pos logit contributions | |
done | +0.048 |
sky | +0.047 |
wrong | +0.046 |
| +0.045 |
am | +0.045 |
Neg logit contributions | |
cess | -0.081 |
ck | -0.077 |
aret | -0.076 |
rily | -0.075 |
eter | -0.075 |
Token | her |
Token ID | 607 |
Feature activation | +2.75e+00 |
Pos logit contributions | |
cess | +0.053 |
charged | +0.050 |
ck | +0.049 |
abbit | +0.048 |
aret | +0.047 |
Neg logit contributions | |
wrong | -0.029 |
done | -0.029 |
made | -0.028 |
missing | -0.027 |
sky | -0.027 |
Token | dad |
Token ID | 9955 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
cess | +0.068 |
charged | +0.063 |
ck | +0.062 |
abbit | +0.060 |
aret | +0.060 |
Neg logit contributions | |
wrong | -0.057 |
done | -0.057 |
best | -0.056 |
made | -0.056 |
missing | -0.055 |
Token | . |
Token ID | 13 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
cess | +0.093 |
charged | +0.085 |
ck | +0.084 |
aret | +0.082 |
abbit | +0.082 |
Neg logit contributions | |
wrong | -0.053 |
done | -0.053 |
made | -0.050 |
sky | -0.050 |
missing | -0.049 |
Token | But |
Token ID | 887 |
Feature activation | +7.42e-01 |
Pos logit contributions | |
cess | +0.015 |
charged | +0.014 |
ck | +0.014 |
aret | +0.013 |
abbit | +0.013 |
Neg logit contributions | |
done | -0.009 |
wrong | -0.009 |
made | -0.009 |
sky | -0.009 |
had | -0.009 |
Token | when |
Token ID | 618 |
Feature activation | -1.68e+00 |
Pos logit contributions | |
cess | +0.020 |
charged | +0.018 |
ck | +0.018 |
aret | +0.018 |
abbit | +0.018 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
sky | -0.013 |
made | -0.013 |
missing | -0.013 |
Token | it |
Token ID | 340 |
Feature activation | -7.47e+00 |
Pos logit contributions | |
wrong | +0.029 |
done | +0.028 |
missing | +0.027 |
sky | +0.026 |
can | +0.026 |
Neg logit contributions | |
cess | -0.047 |
aret | -0.042 |
ck | -0.041 |
charged | -0.041 |
oxy | -0.041 |
Token | was |
Token ID | 373 |
Feature activation | -3.08e+00 |
Pos logit contributions | |
wrong | +0.077 |
sky | +0.068 |
biggest | +0.067 |
best | +0.066 |
done | +0.066 |
Neg logit contributions | |
cess | -0.234 |
ck | -0.226 |
chick | -0.214 |
aret | -0.211 |
abbit | -0.209 |
Token | time |
Token ID | 640 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
am | +0.037 |
can | +0.037 |
boards | +0.037 |
has | +0.036 |
sky | +0.035 |
Neg logit contributions | |
aret | -0.089 |
cess | -0.086 |
ck | -0.086 |
charged | -0.082 |
abbit | -0.082 |
Token | to |
Token ID | 284 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
coolest | +0.078 |
sky | +0.077 |
biggest | +0.076 |
toilet | +0.075 |
loads | +0.073 |
Neg logit contributions | |
aret | -0.064 |
cess | -0.060 |
rily | -0.057 |
fully | -0.056 |
Him | -0.055 |
Token | go |
Token ID | 467 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
cess | +0.060 |
charged | +0.054 |
ck | +0.054 |
aret | +0.053 |
abbit | +0.053 |
Neg logit contributions | |
wrong | -0.034 |
done | -0.033 |
made | -0.032 |
sky | -0.032 |
missing | -0.031 |
Token | , |
Token ID | 11 |
Feature activation | +6.68e-01 |
Pos logit contributions | |
wrong | +0.024 |
done | +0.024 |
boards | +0.024 |
sky | +0.024 |
made | +0.024 |
Neg logit contributions | |
cess | -0.029 |
aret | -0.026 |
charged | -0.024 |
chick | -0.024 |
God | -0.024 |
Token | of |
Token ID | 286 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
cess | +0.016 |
ck | +0.014 |
charged | +0.014 |
aret | +0.014 |
abbit | +0.014 |
Neg logit contributions | |
wrong | -0.007 |
done | -0.007 |
missing | -0.007 |
sky | -0.007 |
made | -0.006 |
Token | fun |
Token ID | 1257 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
cess | +0.041 |
charged | +0.038 |
ck | +0.037 |
aret | +0.037 |
abbit | +0.036 |
Neg logit contributions | |
wrong | -0.024 |
done | -0.024 |
made | -0.023 |
sky | -0.023 |
missing | -0.023 |
Token | games |
Token ID | 1830 |
Feature activation | +5.89e-01 |
Pos logit contributions | |
cess | +0.050 |
ck | +0.045 |
charged | +0.044 |
abbit | +0.043 |
tune | +0.043 |
Neg logit contributions | |
wrong | -0.046 |
done | -0.045 |
missing | -0.044 |
sky | -0.044 |
made | -0.042 |
Token | and |
Token ID | 290 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
cess | +0.015 |
ck | +0.014 |
charged | +0.013 |
tune | +0.013 |
eter | +0.013 |
Neg logit contributions | |
wrong | -0.011 |
done | -0.010 |
missing | -0.010 |
sky | -0.010 |
can | -0.009 |
Token | when |
Token ID | 618 |
Feature activation | -2.06e+00 |
Pos logit contributions | |
cess | +0.081 |
ck | +0.076 |
tune | +0.073 |
charged | +0.073 |
aret | +0.071 |
Neg logit contributions | |
wrong | -0.043 |
missing | -0.038 |
sky | -0.038 |
done | -0.036 |
can | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -7.47e+00 |
Pos logit contributions | |
wrong | +0.037 |
done | +0.036 |
made | +0.035 |
missing | +0.034 |
can | +0.034 |
Neg logit contributions | |
cess | -0.054 |
charged | -0.049 |
ck | -0.049 |
aret | -0.048 |
oxy | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
wrong | +0.093 |
missing | +0.084 |
biggest | +0.079 |
done | +0.076 |
best | +0.075 |
Neg logit contributions | |
cess | -0.203 |
ck | -0.201 |
tune | -0.192 |
feat | -0.188 |
chick | -0.185 |
Token | time |
Token ID | 640 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
am | +0.049 |
can | +0.048 |
has | +0.045 |
boards | +0.044 |
wrong | +0.043 |
Neg logit contributions | |
cess | -0.090 |
aret | -0.090 |
ck | -0.088 |
tune | -0.087 |
ishment | -0.085 |
Token | to |
Token ID | 284 |
Feature activation | +1.13e+00 |
Pos logit contributions | |
sky | +0.075 |
am | +0.074 |
coolest | +0.073 |
toilet | +0.072 |
woods | +0.072 |
Neg logit contributions | |
aret | -0.064 |
rily | -0.060 |
fully | -0.057 |
cess | -0.056 |
.'" | -0.056 |
Token | finish |
Token ID | 5461 |
Feature activation | -1.05e+00 |
Pos logit contributions | |
cess | +0.032 |
ck | +0.030 |
aret | +0.029 |
charged | +0.029 |
oxy | +0.028 |
Neg logit contributions | |
wrong | -0.018 |
done | -0.018 |
sky | -0.017 |
missing | -0.017 |
made | -0.017 |
Token | , |
Token ID | 11 |
Feature activation | +2.02e-01 |
Pos logit contributions | |
wrong | +0.021 |
sky | +0.021 |
done | +0.021 |
can | +0.021 |
made | +0.020 |
Neg logit contributions | |
cess | -0.025 |
charged | -0.022 |
ck | -0.022 |
aret | -0.021 |
rily | -0.021 |
Token | fun |
Token ID | 1257 |
Feature activation | +5.47e-01 |
Pos logit contributions | |
wrong | +0.044 |
done | +0.041 |
sky | +0.041 |
missing | +0.041 |
can | +0.039 |
Neg logit contributions | |
cess | -0.115 |
charged | -0.108 |
ck | -0.107 |
wrists | -0.105 |
oxy | -0.105 |
Token | together |
Token ID | 1978 |
Feature activation | -3.53e-01 |
Pos logit contributions | |
cess | +0.014 |
ck | +0.013 |
charged | +0.013 |
abbit | +0.013 |
aret | +0.013 |
Neg logit contributions | |
wrong | -0.010 |
done | -0.010 |
sky | -0.009 |
missing | -0.009 |
can | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -5.38e-01 |
Pos logit contributions | |
wrong | +0.006 |
done | +0.006 |
sky | +0.006 |
missing | +0.006 |
made | +0.006 |
Neg logit contributions | |
cess | -0.010 |
ck | -0.009 |
charged | -0.009 |
abbit | -0.009 |
aret | -0.008 |
Token | But |
Token ID | 887 |
Feature activation | +1.75e-01 |
Pos logit contributions | |
done | +0.013 |
wrong | +0.012 |
am | +0.012 |
missing | +0.011 |
can | +0.011 |
Neg logit contributions | |
cess | -0.011 |
ishment | -0.010 |
aret | -0.010 |
eter | -0.009 |
wrists | -0.009 |
Token | when |
Token ID | 618 |
Feature activation | -2.15e+00 |
Pos logit contributions | |
cess | +0.004 |
ck | +0.003 |
aret | +0.003 |
ishment | +0.003 |
charged | +0.003 |
Neg logit contributions | |
sky | -0.004 |
boards | -0.004 |
done | -0.004 |
wrong | -0.004 |
can | -0.004 |
Token | it |
Token ID | 340 |
Feature activation | -7.46e+00 |
Pos logit contributions | |
boards | +0.046 |
wrong | +0.045 |
sky | +0.045 |
done | +0.044 |
missing | +0.042 |
Neg logit contributions | |
cess | -0.046 |
ck | -0.040 |
aret | -0.038 |
feat | -0.037 |
cle | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | -3.51e+00 |
Pos logit contributions | |
wrong | +0.074 |
sky | +0.069 |
best | +0.067 |
done | +0.063 |
biggest | +0.063 |
Neg logit contributions | |
cess | -0.232 |
ck | -0.227 |
tune | -0.209 |
charged | -0.208 |
oing | -0.207 |
Token | time |
Token ID | 640 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
can | +0.047 |
am | +0.047 |
has | +0.046 |
boards | +0.044 |
sky | +0.042 |
Neg logit contributions | |
cess | -0.093 |
ck | -0.092 |
aret | -0.090 |
charged | -0.089 |
abbit | -0.088 |
Token | for |
Token ID | 329 |
Feature activation | +2.52e-01 |
Pos logit contributions | |
sky | +0.079 |
coolest | +0.076 |
biggest | +0.075 |
woods | +0.074 |
most | +0.073 |
Neg logit contributions | |
aret | -0.061 |
cess | -0.057 |
fully | -0.055 |
rily | -0.054 |
.'" | -0.053 |
Token | Tim |
Token ID | 5045 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
cess | +0.006 |
ck | +0.006 |
charged | +0.006 |
aret | +0.006 |
oxy | +0.005 |
Neg logit contributions | |
wrong | -0.005 |
can | -0.005 |
sky | -0.005 |
has | -0.005 |
done | -0.005 |
Token | my |
Token ID | 1820 |
Feature activation | -3.25e-01 |
Pos logit contributions | |
cess | +0.083 |
charged | +0.076 |
ck | +0.075 |
aret | +0.073 |
rily | +0.073 |
Neg logit contributions | |
done | -0.044 |
wrong | -0.043 |
made | -0.042 |
had | -0.040 |
missing | -0.040 |
Token | much |
Token ID | 881 |
Feature activation | -3.67e-01 |
Pos logit contributions | |
wrong | +0.003 |
missing | +0.003 |
sky | +0.003 |
done | +0.002 |
can | +0.002 |
Neg logit contributions | |
ck | -0.005 |
cess | -0.005 |
charged | -0.005 |
oxy | -0.005 |
abbit | -0.005 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.24e+00 |
Pos logit contributions | |
wrong | +0.006 |
missing | +0.006 |
has | +0.006 |
sky | +0.006 |
can | +0.006 |
Neg logit contributions | |
feat | -0.008 |
ck | -0.008 |
tune | -0.007 |
eter | -0.007 |
tunes | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +2.25e+00 |
Pos logit contributions | |
cess | +0.030 |
ck | +0.029 |
feat | +0.029 |
tune | +0.028 |
tunes | +0.027 |
Neg logit contributions | |
wrong | -0.023 |
sky | -0.022 |
done | -0.022 |
missing | -0.021 |
can | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +2.59e+00 |
Pos logit contributions | |
cess | +0.061 |
ck | +0.055 |
charged | +0.054 |
aret | +0.054 |
tune | +0.053 |
Neg logit contributions | |
wrong | -0.040 |
done | -0.038 |
sky | -0.037 |
missing | -0.036 |
can | -0.035 |
Token | when |
Token ID | 618 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
cess | +0.058 |
aret | +0.050 |
ck | +0.049 |
charged | +0.049 |
tune | +0.048 |
Neg logit contributions | |
wrong | -0.058 |
missing | -0.056 |
sky | -0.056 |
done | -0.055 |
can | -0.052 |
Token | it |
Token ID | 340 |
Feature activation | -7.45e+00 |
Pos logit contributions | |
wrong | +0.028 |
done | +0.028 |
missing | +0.027 |
sky | +0.027 |
made | +0.026 |
Neg logit contributions | |
cess | -0.039 |
aret | -0.035 |
charged | -0.034 |
oxy | -0.034 |
ck | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | -3.05e+00 |
Pos logit contributions | |
wrong | +0.098 |
missing | +0.084 |
seeker | +0.083 |
biggest | +0.083 |
done | +0.080 |
Neg logit contributions | |
ck | -0.201 |
cess | -0.199 |
feat | -0.186 |
ie | -0.181 |
chick | -0.179 |
Token | time |
Token ID | 640 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
am | +0.046 |
boards | +0.044 |
can | +0.043 |
has | +0.042 |
wrong | +0.040 |
Neg logit contributions | |
aret | -0.081 |
ck | -0.080 |
cess | -0.079 |
rily | -0.076 |
ishment | -0.076 |
Token | to |
Token ID | 284 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
coolest | +0.075 |
sky | +0.075 |
biggest | +0.074 |
am | +0.073 |
woods | +0.071 |
Neg logit contributions | |
aret | -0.056 |
rily | -0.050 |
fully | -0.049 |
cess | -0.049 |
heads | -0.047 |
Token | go |
Token ID | 467 |
Feature activation | -1.30e+00 |
Pos logit contributions | |
cess | +0.052 |
ck | +0.048 |
charged | +0.047 |
aret | +0.047 |
abbit | +0.045 |
Neg logit contributions | |
wrong | -0.029 |
done | -0.028 |
sky | -0.027 |
missing | -0.026 |
made | -0.026 |
Token | home |
Token ID | 1363 |
Feature activation | -5.94e-03 |
Pos logit contributions | |
wrong | +0.029 |
boards | +0.029 |
done | +0.028 |
made | +0.028 |
sky | +0.027 |
Neg logit contributions | |
cess | -0.024 |
aret | -0.021 |
ck | -0.021 |
charged | -0.021 |
feat | -0.020 |
Token | and |
Token ID | 290 |
Feature activation | +2.69e+00 |
Pos logit contributions | |
cess | +0.080 |
aret | +0.073 |
charged | +0.073 |
oxy | +0.071 |
cle | +0.071 |
Neg logit contributions | |
all | -0.032 |
on | -0.032 |
made | -0.032 |
done | -0.030 |
wrong | -0.030 |
Token | seek |
Token ID | 5380 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
cess | +0.057 |
charged | +0.049 |
ck | +0.048 |
aret | +0.048 |
abbit | +0.047 |
Neg logit contributions | |
done | -0.066 |
wrong | -0.066 |
made | -0.064 |
sky | -0.063 |
had | -0.063 |
Token | , |
Token ID | 11 |
Feature activation | +9.76e-01 |
Pos logit contributions | |
cess | +0.035 |
charged | +0.032 |
ck | +0.032 |
abbit | +0.031 |
aret | +0.031 |
Neg logit contributions | |
wrong | -0.015 |
done | -0.015 |
sky | -0.014 |
missing | -0.014 |
made | -0.014 |
Token | but |
Token ID | 475 |
Feature activation | -3.35e-01 |
Pos logit contributions | |
cess | +0.028 |
ck | +0.025 |
charged | +0.025 |
aret | +0.024 |
abbit | +0.024 |
Neg logit contributions | |
wrong | -0.017 |
done | -0.016 |
sky | -0.015 |
missing | -0.015 |
made | -0.015 |
Token | when |
Token ID | 618 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
done | +0.007 |
wrong | +0.007 |
sky | +0.007 |
boards | +0.007 |
missing | +0.007 |
Neg logit contributions | |
cess | -0.008 |
ck | -0.007 |
charged | -0.007 |
abbit | -0.007 |
ishment | -0.006 |
Token | it |
Token ID | 340 |
Feature activation | -7.41e+00 |
Pos logit contributions | |
wrong | +0.034 |
done | +0.034 |
missing | +0.032 |
sky | +0.031 |
made | +0.031 |
Neg logit contributions | |
cess | -0.055 |
ck | -0.049 |
charged | -0.049 |
oxy | -0.048 |
abbit | -0.048 |
Token | was |
Token ID | 373 |
Feature activation | -3.49e+00 |
Pos logit contributions | |
wrong | +0.077 |
sky | +0.070 |
done | +0.068 |
missing | +0.068 |
best | +0.067 |
Neg logit contributions | |
cess | -0.226 |
ck | -0.223 |
abbit | -0.202 |
chick | -0.202 |
feat | -0.201 |
Token | P |
Token ID | 350 |
Feature activation | +7.17e-01 |
Pos logit contributions | |
can | +0.044 |
am | +0.044 |
has | +0.043 |
boards | +0.041 |
sky | +0.040 |
Neg logit contributions | |
cess | -0.095 |
aret | -0.093 |
ck | -0.092 |
charged | -0.089 |
abbit | -0.089 |
Token | andy |
Token ID | 10757 |
Feature activation | -1.41e+00 |
Pos logit contributions | |
cess | +0.012 |
charged | +0.010 |
ck | +0.010 |
abbit | +0.010 |
oxy | +0.010 |
Neg logit contributions | |
wrong | -0.020 |
done | -0.020 |
sky | -0.020 |
made | -0.020 |
missing | -0.020 |
Token | 's |
Token ID | 338 |
Feature activation | -2.45e+00 |
Pos logit contributions | |
wrong | +0.016 |
done | +0.016 |
sky | +0.016 |
missing | +0.014 |
can | +0.014 |
Neg logit contributions | |
cess | -0.046 |
ck | -0.042 |
charged | -0.042 |
oxy | -0.041 |
abbit | -0.041 |
Token | turn |
Token ID | 1210 |
Feature activation | -3.27e+00 |
Pos logit contributions | |
wrong | +0.050 |
done | +0.049 |
made | +0.048 |
sky | +0.047 |
missing | +0.046 |
Neg logit contributions | |
cess | -0.061 |
charged | -0.055 |
ck | -0.054 |
aret | -0.053 |
abbit | -0.053 |
Token | much |
Token ID | 881 |
Feature activation | -5.25e-01 |
Pos logit contributions | |
cess | +0.017 |
ck | +0.016 |
charged | +0.016 |
oxy | +0.016 |
abbit | +0.015 |
Neg logit contributions | |
wrong | -0.007 |
done | -0.007 |
sky | -0.007 |
missing | -0.006 |
made | -0.006 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
wrong | +0.008 |
missing | +0.008 |
has | +0.008 |
can | +0.008 |
had | +0.008 |
Neg logit contributions | |
ck | -0.013 |
feat | -0.013 |
cess | -0.012 |
tune | -0.012 |
tunes | -0.011 |
Token | . |
Token ID | 13 |
Feature activation | +2.44e-01 |
Pos logit contributions | |
cess | +0.028 |
ck | +0.026 |
charged | +0.025 |
tune | +0.025 |
feat | +0.025 |
Neg logit contributions | |
wrong | -0.020 |
done | -0.020 |
sky | -0.019 |
can | -0.019 |
missing | -0.019 |
Token | But |
Token ID | 887 |
Feature activation | +4.23e-01 |
Pos logit contributions | |
cess | +0.007 |
aret | +0.006 |
abbit | +0.006 |
charged | +0.006 |
oxy | +0.006 |
Neg logit contributions | |
done | -0.004 |
wrong | -0.004 |
made | -0.004 |
missing | -0.004 |
best | -0.004 |
Token | when |
Token ID | 618 |
Feature activation | -2.43e+00 |
Pos logit contributions | |
cess | +0.010 |
ck | +0.009 |
aret | +0.008 |
charged | +0.008 |
ishment | +0.008 |
Neg logit contributions | |
done | -0.009 |
sky | -0.009 |
wrong | -0.009 |
boards | -0.009 |
can | -0.009 |
Token | it |
Token ID | 340 |
Feature activation | -7.41e+00 |
Pos logit contributions | |
boards | +0.047 |
done | +0.045 |
wrong | +0.044 |
sky | +0.043 |
missing | +0.041 |
Neg logit contributions | |
cess | -0.055 |
ck | -0.049 |
aret | -0.048 |
charged | -0.047 |
chick | -0.045 |
Token | was |
Token ID | 373 |
Feature activation | -3.26e+00 |
Pos logit contributions | |
wrong | +0.083 |
best | +0.072 |
biggest | +0.070 |
boards | +0.069 |
missing | +0.069 |
Neg logit contributions | |
ck | -0.214 |
cess | -0.212 |
tune | -0.198 |
aunted | -0.193 |
ie | -0.191 |
Token | time |
Token ID | 640 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
can | +0.050 |
am | +0.049 |
boards | +0.047 |
has | +0.047 |
is | +0.043 |
Neg logit contributions | |
cess | -0.078 |
ck | -0.078 |
aret | -0.078 |
tune | -0.076 |
heads | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
coolest | +0.083 |
biggest | +0.082 |
sky | +0.081 |
boards | +0.079 |
loads | +0.078 |
Neg logit contributions | |
aret | -0.056 |
cess | -0.055 |
fully | -0.050 |
ck | -0.049 |
heads | -0.048 |
Token | go |
Token ID | 467 |
Feature activation | -1.53e+00 |
Pos logit contributions | |
cess | +0.059 |
charged | +0.054 |
ck | +0.054 |
aret | +0.053 |
abbit | +0.052 |
Neg logit contributions | |
wrong | -0.027 |
done | -0.027 |
made | -0.026 |
sky | -0.025 |
missing | -0.025 |
Token | home |
Token ID | 1363 |
Feature activation | -5.62e-01 |
Pos logit contributions | |
boards | +0.034 |
made | +0.032 |
sky | +0.032 |
wrong | +0.032 |
done | +0.032 |
Neg logit contributions | |
cess | -0.028 |
ck | -0.025 |
ie | -0.024 |
aret | -0.023 |
feat | -0.023 |
Token | |
Token ID | 220 |
Feature activation | -2.38e+00 |
Pos logit contributions | |
wrong | +0.017 |
done | +0.016 |
missing | +0.016 |
made | +0.014 |
best | +0.014 |
Neg logit contributions | |
cess | -0.027 |
aret | -0.024 |
oxy | -0.024 |
ck | -0.024 |
abbit | -0.024 |
Token |
|
Token ID | 198 |
Feature activation | -2.69e-01 |
Pos logit contributions | |
wrong | +0.030 |
done | +0.029 |
made | +0.028 |
sky | +0.027 |
missing | +0.027 |
Neg logit contributions | |
cess | -0.078 |
charged | -0.072 |
ck | -0.071 |
aret | -0.070 |
abbit | -0.070 |
Token |
|
Token ID | 198 |
Feature activation | +1.18e+00 |
Pos logit contributions | |
wrong | +0.003 |
done | +0.003 |
made | +0.003 |
sky | +0.003 |
missing | +0.003 |
Neg logit contributions | |
cess | -0.009 |
charged | -0.008 |
ck | -0.008 |
aret | -0.008 |
abbit | -0.008 |
Token | But |
Token ID | 1537 |
Feature activation | -4.50e-02 |
Pos logit contributions | |
cess | +0.027 |
ck | +0.024 |
charged | +0.024 |
oxy | +0.023 |
aret | +0.023 |
Neg logit contributions | |
wrong | -0.027 |
done | -0.026 |
missing | -0.025 |
sky | -0.025 |
made | -0.025 |
Token | when |
Token ID | 618 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
wrong | +0.001 |
done | +0.001 |
sky | +0.001 |
missing | +0.001 |
can | +0.001 |
Neg logit contributions | |
cess | -0.001 |
ck | -0.001 |
charged | -0.001 |
aret | -0.001 |
abbit | -0.001 |
Token | it |
Token ID | 340 |
Feature activation | -7.40e+00 |
Pos logit contributions | |
wrong | +0.043 |
done | +0.042 |
missing | +0.041 |
sky | +0.040 |
made | +0.040 |
Neg logit contributions | |
cess | -0.058 |
ck | -0.052 |
charged | -0.051 |
aret | -0.050 |
oxy | -0.050 |
Token | was |
Token ID | 373 |
Feature activation | -3.40e+00 |
Pos logit contributions | |
wrong | +0.084 |
best | +0.071 |
missing | +0.066 |
done | +0.066 |
biggest | +0.065 |
Neg logit contributions | |
cess | -0.236 |
ck | -0.227 |
tune | -0.212 |
charged | -0.209 |
oing | -0.207 |
Token | time |
Token ID | 640 |
Feature activation | -3.42e+00 |
Pos logit contributions | |
am | +0.048 |
can | +0.046 |
has | +0.044 |
boards | +0.043 |
wrong | +0.040 |
Neg logit contributions | |
aret | -0.088 |
cess | -0.088 |
ck | -0.086 |
tune | -0.085 |
charged | -0.084 |
Token | to |
Token ID | 284 |
Feature activation | +1.67e+00 |
Pos logit contributions | |
coolest | +0.078 |
biggest | +0.077 |
sky | +0.076 |
tallest | +0.073 |
most | +0.073 |
Neg logit contributions | |
aret | -0.056 |
cess | -0.052 |
fully | -0.052 |
rily | -0.049 |
heads | -0.045 |
Token | go |
Token ID | 467 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
cess | +0.050 |
charged | +0.046 |
ck | +0.045 |
aret | +0.045 |
abbit | +0.044 |
Neg logit contributions | |
wrong | -0.026 |
done | -0.025 |
made | -0.024 |
sky | -0.024 |
missing | -0.024 |
Token | home |
Token ID | 1363 |
Feature activation | -4.85e-01 |
Pos logit contributions | |
boards | +0.034 |
made | +0.033 |
wrong | +0.033 |
done | +0.033 |
sky | +0.032 |
Neg logit contributions | |
cess | -0.029 |
aret | -0.023 |
ck | -0.023 |
charged | -0.023 |
someday | -0.023 |
Token | with |
Token ID | 351 |
Feature activation | +5.70e-01 |
Pos logit contributions | |
done | +0.006 |
wrong | +0.006 |
can | +0.006 |
made | +0.006 |
had | +0.006 |
Neg logit contributions | |
cess | -0.013 |
ck | -0.012 |
rily | -0.012 |
charged | -0.012 |
aret | -0.011 |
Token | joy |
Token ID | 8716 |
Feature activation | +1.99e+00 |
Pos logit contributions | |
cess | +0.017 |
charged | +0.015 |
ck | +0.015 |
aret | +0.015 |
abbit | +0.015 |
Neg logit contributions | |
wrong | -0.009 |
done | -0.009 |
made | -0.009 |
sky | -0.009 |
missing | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -4.97e-02 |
Pos logit contributions | |
cess | +0.054 |
charged | +0.049 |
ck | +0.048 |
aret | +0.048 |
abbit | +0.047 |
Neg logit contributions | |
wrong | -0.036 |
done | -0.035 |
sky | -0.034 |
made | -0.033 |
can | -0.033 |
Token | But |
Token ID | 887 |
Feature activation | +3.00e-01 |
Pos logit contributions | |
wrong | +0.001 |
done | +0.001 |
missing | +0.001 |
made | +0.001 |
on | +0.001 |
Neg logit contributions | |
cess | -0.001 |
aret | -0.001 |
eter | -0.001 |
charged | -0.001 |
abbit | -0.001 |
Token | when |
Token ID | 618 |
Feature activation | -2.35e+00 |
Pos logit contributions | |
cess | +0.007 |
ck | +0.007 |
charged | +0.007 |
aret | +0.006 |
abbit | +0.006 |
Neg logit contributions | |
wrong | -0.006 |
done | -0.006 |
sky | -0.006 |
can | -0.006 |
missing | -0.005 |
Token | it |
Token ID | 340 |
Feature activation | -7.37e+00 |
Pos logit contributions | |
wrong | +0.042 |
boards | +0.040 |
has | +0.038 |
done | +0.038 |
sky | +0.038 |
Neg logit contributions | |
cess | -0.058 |
ck | -0.054 |
feat | -0.053 |
charged | -0.051 |
aret | -0.050 |
Token | came |
Token ID | 1625 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
wrong | +0.083 |
biggest | +0.079 |
sky | +0.077 |
best | +0.077 |
living | +0.072 |
Neg logit contributions | |
cess | -0.211 |
ck | -0.208 |
aunted | -0.186 |
eter | -0.186 |
bug | -0.185 |
Token | time |
Token ID | 640 |
Feature activation | -3.66e+00 |
Pos logit contributions | |
done | +0.032 |
wrong | +0.031 |
made | +0.030 |
had | +0.028 |
missing | +0.028 |
Neg logit contributions | |
cess | -0.082 |
charged | -0.075 |
abbit | -0.074 |
aret | -0.073 |
ck | -0.073 |
Token | for |
Token ID | 329 |
Feature activation | -2.95e-02 |
Pos logit contributions | |
coolest | +0.067 |
biggest | +0.067 |
sky | +0.066 |
am | +0.065 |
most | +0.064 |
Neg logit contributions | |
cess | -0.078 |
aret | -0.074 |
rily | -0.071 |
oing | -0.066 |
heads | -0.065 |
Token | Suz |
Token ID | 22691 |
Feature activation | +3.60e+00 |
Pos logit contributions | |
wrong | +0.000 |
can | +0.000 |
has | +0.000 |
sky | +0.000 |
missing | +0.000 |
Neg logit contributions | |
cess | -0.001 |
ck | -0.001 |
charged | -0.001 |
aret | -0.001 |
oxy | -0.001 |
Token | ie |
Token ID | 494 |
Feature activation | +3.08e-01 |
Pos logit contributions | |
cess | +0.045 |
charged | +0.034 |
ck | +0.032 |
abbit | +0.032 |
oxy | +0.031 |
Neg logit contributions | |
wrong | -0.122 |
done | -0.122 |
made | -0.118 |
missing | -0.117 |
had | -0.116 |
Token |
|
Token ID | 198 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
done | +0.000 |
wrong | +0.000 |
sky | +0.000 |
had | +0.000 |
made | +0.000 |
Neg logit contributions | |
cess | -0.001 |
charged | -0.001 |
ck | -0.001 |
ishment | -0.001 |
eter | -0.001 |
Token | But |
Token ID | 1537 |
Feature activation | +1.64e-01 |
Pos logit contributions | |
cess | +0.038 |
charged | +0.036 |
ishment | +0.034 |
abbit | +0.034 |
eter | +0.033 |
Neg logit contributions | |
done | -0.028 |
had | -0.027 |
sky | -0.027 |
made | -0.027 |
wrong | -0.026 |
Token | then |
Token ID | 788 |
Feature activation | -1.15e+00 |
Pos logit contributions | |
cess | +0.004 |
ck | +0.004 |
charged | +0.004 |
aret | +0.004 |
oxy | +0.003 |
Neg logit contributions | |
wrong | -0.003 |
done | -0.003 |
sky | -0.003 |
missing | -0.003 |
can | -0.003 |
Token | , |
Token ID | 11 |
Feature activation | -1.19e+00 |
Pos logit contributions | |
wrong | +0.021 |
missing | +0.020 |
done | +0.020 |
sky | +0.019 |
can | +0.019 |
Neg logit contributions | |
cess | -0.031 |
ck | -0.027 |
aret | -0.027 |
oxy | -0.026 |
charged | -0.026 |
Token | when |
Token ID | 618 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
wrong | +0.026 |
done | +0.025 |
missing | +0.025 |
can | +0.024 |
sky | +0.024 |
Neg logit contributions | |
cess | -0.028 |
ck | -0.024 |
aret | -0.024 |
oxy | -0.023 |
charged | -0.023 |
Token | it |
Token ID | 340 |
Feature activation | -7.36e+00 |
Pos logit contributions | |
wrong | +0.044 |
done | +0.042 |
missing | +0.042 |
can | +0.041 |
sky | +0.041 |
Neg logit contributions | |
cess | -0.051 |
ck | -0.044 |
aret | -0.044 |
charged | -0.043 |
oxy | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | -3.50e+00 |
Pos logit contributions | |
wrong | +0.094 |
best | +0.082 |
biggest | +0.079 |
missing | +0.076 |
done | +0.075 |
Neg logit contributions | |
ck | -0.213 |
cess | -0.211 |
chick | -0.193 |
tune | -0.193 |
feat | -0.192 |
Token | time |
Token ID | 640 |
Feature activation | -3.70e+00 |
Pos logit contributions | |
can | +0.047 |
has | +0.046 |
am | +0.046 |
boards | +0.044 |
is | +0.041 |
Neg logit contributions | |
cess | -0.096 |
ck | -0.095 |
aret | -0.093 |
tune | -0.090 |
ishment | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
coolest | +0.085 |
biggest | +0.083 |
tallest | +0.079 |
sky | +0.079 |
loads | +0.079 |
Neg logit contributions | |
aret | -0.060 |
fully | -0.056 |
rily | -0.053 |
cess | -0.051 |
heads | -0.050 |
Token | take |
Token ID | 1011 |
Feature activation | -1.04e+00 |
Pos logit contributions | |
cess | +0.054 |
charged | +0.051 |
abbit | +0.049 |
ck | +0.049 |
aret | +0.049 |
Neg logit contributions | |
done | -0.024 |
wrong | -0.024 |
made | -0.023 |
sky | -0.022 |
missing | -0.022 |
Token | the |
Token ID | 262 |
Feature activation | -1.46e+00 |
Pos logit contributions | |
wrong | +0.014 |
done | +0.013 |
made | +0.013 |
sky | +0.012 |
missing | +0.012 |
Neg logit contributions | |
cess | -0.034 |
charged | -0.031 |
ck | -0.031 |
aret | -0.030 |
abbit | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | +2.81e-01 |
Pos logit contributions | |
done | +0.035 |
wrong | +0.035 |
sky | +0.034 |
missing | +0.033 |
made | +0.033 |
Neg logit contributions | |
cess | -0.051 |
ck | -0.046 |
charged | -0.046 |
aret | -0.045 |
eter | -0.044 |
Token |
|
Token ID | 198 |
Feature activation | -1.77e-01 |
Pos logit contributions | |
cess | +0.009 |
charged | +0.008 |
ck | +0.008 |
aret | +0.008 |
abbit | +0.008 |
Neg logit contributions | |
wrong | -0.004 |
done | -0.004 |
made | -0.004 |
missing | -0.004 |
sky | -0.004 |
Token |
|
Token ID | 198 |
Feature activation | +1.43e+00 |
Pos logit contributions | |
wrong | +0.002 |
done | +0.002 |
made | +0.002 |
missing | +0.002 |
sky | +0.002 |
Neg logit contributions | |
cess | -0.006 |
charged | -0.005 |
ck | -0.005 |
aret | -0.005 |
abbit | -0.005 |
Token | But |
Token ID | 1537 |
Feature activation | -1.05e-01 |
Pos logit contributions | |
cess | +0.035 |
charged | +0.032 |
ck | +0.031 |
aret | +0.031 |
abbit | +0.030 |
Neg logit contributions | |
done | -0.029 |
wrong | -0.029 |
made | -0.029 |
sky | -0.028 |
had | -0.028 |
Token | when |
Token ID | 618 |
Feature activation | -1.88e+00 |
Pos logit contributions | |
wrong | +0.002 |
done | +0.002 |
sky | +0.002 |
missing | +0.002 |
can | +0.002 |
Neg logit contributions | |
cess | -0.003 |
charged | -0.002 |
aret | -0.002 |
ck | -0.002 |
abbit | -0.002 |
Token | it |
Token ID | 340 |
Feature activation | -7.36e+00 |
Pos logit contributions | |
wrong | +0.037 |
done | +0.036 |
missing | +0.035 |
sky | +0.035 |
can | +0.034 |
Neg logit contributions | |
cess | -0.047 |
aret | -0.041 |
charged | -0.039 |
cle | -0.039 |
ck | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -3.59e+00 |
Pos logit contributions | |
wrong | +0.083 |
best | +0.072 |
sky | +0.070 |
missing | +0.070 |
done | +0.070 |
Neg logit contributions | |
cess | -0.225 |
ck | -0.219 |
aunted | -0.202 |
feat | -0.201 |
eter | -0.200 |
Token | time |
Token ID | 640 |
Feature activation | -3.77e+00 |
Pos logit contributions | |
can | +0.046 |
am | +0.046 |
has | +0.045 |
sky | +0.043 |
boards | +0.041 |
Neg logit contributions | |
aret | -0.101 |
cess | -0.100 |
ck | -0.099 |
ishment | -0.094 |
heads | -0.093 |
Token | for |
Token ID | 329 |
Feature activation | +3.93e-01 |
Pos logit contributions | |
coolest | +0.084 |
sky | +0.084 |
biggest | +0.082 |
woods | +0.081 |
tallest | +0.080 |
Neg logit contributions | |
aret | -0.067 |
fully | -0.057 |
rily | -0.057 |
cess | -0.056 |
.'" | -0.053 |
Token | the |
Token ID | 262 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
cess | +0.010 |
ck | +0.009 |
charged | +0.009 |
aret | +0.009 |
oxy | +0.009 |
Neg logit contributions | |
wrong | -0.008 |
done | -0.007 |
can | -0.007 |
sky | -0.007 |
made | -0.007 |
Token | match |
Token ID | 2872 |
Feature activation | -3.33e+00 |
Pos logit contributions | |
done | +0.028 |
made | +0.027 |
wrong | +0.026 |
sky | +0.026 |
can | +0.026 |
Neg logit contributions | |
cess | -0.049 |
charged | -0.045 |
ck | -0.044 |
tune | -0.044 |
abbit | -0.043 |
Token | playing |
Token ID | 2712 |
Feature activation | +5.45e-02 |
Pos logit contributions | |
cess | +0.038 |
charged | +0.035 |
ck | +0.034 |
aret | +0.034 |
abbit | +0.034 |
Neg logit contributions | |
wrong | -0.020 |
done | -0.019 |
made | -0.018 |
sky | -0.018 |
missing | -0.018 |
Token | together |
Token ID | 1978 |
Feature activation | +1.10e+00 |
Pos logit contributions | |
cess | +0.002 |
charged | +0.001 |
ck | +0.001 |
aret | +0.001 |
abbit | +0.001 |
Neg logit contributions | |
wrong | -0.001 |
done | -0.001 |
made | -0.001 |
sky | -0.001 |
missing | -0.001 |
Token | . |
Token ID | 13 |
Feature activation | -5.26e-01 |
Pos logit contributions | |
cess | +0.033 |
charged | +0.030 |
ck | +0.029 |
aret | +0.029 |
abbit | +0.029 |
Neg logit contributions | |
wrong | -0.017 |
done | -0.016 |
sky | -0.015 |
made | -0.015 |
can | -0.015 |
Token | But |
Token ID | 887 |
Feature activation | +1.47e-01 |
Pos logit contributions | |
wrong | +0.013 |
done | +0.012 |
missing | +0.012 |
made | +0.011 |
best | +0.011 |
Neg logit contributions | |
cess | -0.011 |
aret | -0.009 |
eter | -0.009 |
abbit | -0.009 |
oxy | -0.009 |
Token | when |
Token ID | 618 |
Feature activation | -2.52e+00 |
Pos logit contributions | |
cess | +0.003 |
ck | +0.003 |
aret | +0.003 |
cle | +0.003 |
charged | +0.002 |
Neg logit contributions | |
boards | -0.003 |
sky | -0.003 |
done | -0.003 |
wrong | -0.003 |
am | -0.003 |
Token | it |
Token ID | 340 |
Feature activation | -7.35e+00 |
Pos logit contributions | |
boards | +0.051 |
sky | +0.050 |
wrong | +0.050 |
done | +0.049 |
missing | +0.048 |
Neg logit contributions | |
cess | -0.054 |
ck | -0.048 |
aret | -0.046 |
charged | -0.046 |
cle | -0.046 |
Token | was |
Token ID | 373 |
Feature activation | -3.21e+00 |
Pos logit contributions | |
wrong | +0.078 |
best | +0.068 |
biggest | +0.067 |
sky | +0.066 |
done | +0.065 |
Neg logit contributions | |
cess | -0.226 |
ck | -0.214 |
oing | -0.197 |
feat | -0.197 |
aret | -0.196 |
Token | time |
Token ID | 640 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
am | +0.040 |
can | +0.039 |
has | +0.038 |
wrong | +0.036 |
boards | +0.034 |
Neg logit contributions | |
cess | -0.094 |
aret | -0.092 |
ck | -0.089 |
tune | -0.087 |
charged | -0.086 |
Token | for |
Token ID | 329 |
Feature activation | +4.71e-01 |
Pos logit contributions | |
sky | +0.080 |
coolest | +0.077 |
biggest | +0.076 |
loads | +0.074 |
tallest | +0.074 |
Neg logit contributions | |
aret | -0.059 |
fully | -0.057 |
cess | -0.054 |
rily | -0.052 |
heads | -0.049 |
Token | grandma |
Token ID | 49890 |
Feature activation | +2.67e+00 |
Pos logit contributions | |
cess | +0.013 |
charged | +0.011 |
ck | +0.011 |
aret | +0.011 |
oxy | +0.011 |
Neg logit contributions | |
wrong | -0.008 |
done | -0.008 |
can | -0.008 |
sky | -0.008 |
made | -0.008 |
Token | to |
Token ID | 284 |
Feature activation | +2.10e+00 |
Pos logit contributions | |
cess | +0.085 |
charged | +0.077 |
ck | +0.077 |
aret | +0.076 |
oxy | +0.076 |
Neg logit contributions | |
wrong | -0.035 |
done | -0.034 |
sky | -0.034 |
made | -0.032 |
missing | -0.032 |
Token | all |
Token ID | 477 |
Feature activation | +6.22e-01 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.013 |
ck | +0.013 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
wrong | -0.005 |
done | -0.005 |
made | -0.005 |
sky | -0.005 |
missing | -0.005 |
Token | day |
Token ID | 1110 |
Feature activation | -1.01e+00 |
Pos logit contributions | |
cess | +0.021 |
charged | +0.020 |
ck | +0.020 |
aret | +0.020 |
abbit | +0.019 |
Neg logit contributions | |
wrong | -0.007 |
done | -0.006 |
missing | -0.006 |
sky | -0.006 |
made | -0.006 |
Token | . |
Token ID | 13 |
Feature activation | -7.30e-01 |
Pos logit contributions | |
wrong | +0.015 |
done | +0.014 |
made | +0.014 |
sky | +0.013 |
missing | +0.013 |
Neg logit contributions | |
cess | -0.031 |
charged | -0.029 |
ck | -0.028 |
aret | -0.028 |
abbit | -0.028 |
Token | But |
Token ID | 887 |
Feature activation | +5.37e-03 |
Pos logit contributions | |
wrong | +0.014 |
done | +0.014 |
missing | +0.014 |
made | +0.014 |
sky | +0.013 |
Neg logit contributions | |
cess | -0.019 |
aret | -0.016 |
charged | -0.016 |
abbit | -0.016 |
ck | -0.016 |
Token | when |
Token ID | 618 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
cess | +0.000 |
ck | +0.000 |
aret | +0.000 |
charged | +0.000 |
cle | +0.000 |
Neg logit contributions | |
sky | -0.000 |
done | -0.000 |
boards | -0.000 |
wrong | -0.000 |
missing | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | -7.33e+00 |
Pos logit contributions | |
done | +0.039 |
wrong | +0.038 |
sky | +0.037 |
missing | +0.036 |
boards | +0.035 |
Neg logit contributions | |
cess | -0.045 |
aret | -0.039 |
ck | -0.038 |
charged | -0.038 |
cle | -0.037 |
Token | was |
Token ID | 373 |
Feature activation | -3.25e+00 |
Pos logit contributions | |
wrong | +0.073 |
missing | +0.063 |
best | +0.062 |
done | +0.061 |
sky | +0.061 |
Neg logit contributions | |
cess | -0.234 |
ck | -0.227 |
charged | -0.208 |
aret | -0.207 |
tune | -0.206 |
Token | time |
Token ID | 640 |
Feature activation | -3.56e+00 |
Pos logit contributions | |
can | +0.037 |
am | +0.036 |
has | +0.036 |
wrong | +0.035 |
had | +0.034 |
Neg logit contributions | |
cess | -0.099 |
aret | -0.096 |
ck | -0.095 |
charged | -0.091 |
ishment | -0.090 |
Token | for |
Token ID | 329 |
Feature activation | +5.21e-01 |
Pos logit contributions | |
sky | +0.076 |
coolest | +0.074 |
biggest | +0.073 |
most | +0.073 |
tallest | +0.070 |
Neg logit contributions | |
aret | -0.063 |
cess | -0.057 |
fully | -0.054 |
Him | -0.053 |
.'" | -0.053 |
Token | Max |
Token ID | 5436 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.012 |
ck | +0.012 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
wrong | -0.010 |
done | -0.010 |
made | -0.009 |
sky | -0.009 |
missing | -0.009 |
Token | to |
Token ID | 284 |
Feature activation | +1.72e+00 |
Pos logit contributions | |
cess | +0.033 |
charged | +0.030 |
ck | +0.030 |
aret | +0.030 |
abbit | +0.029 |
Neg logit contributions | |
wrong | -0.012 |
done | -0.012 |
made | -0.011 |
missing | -0.011 |
sky | -0.011 |
Token | for |
Token ID | 329 |
Feature activation | +1.38e+00 |
Pos logit contributions | |
done | +0.015 |
wrong | +0.015 |
sky | +0.015 |
missing | +0.015 |
made | +0.014 |
Neg logit contributions | |
cess | -0.022 |
ck | -0.020 |
charged | -0.020 |
tune | -0.019 |
abbit | -0.019 |
Token | hours |
Token ID | 2250 |
Feature activation | +7.76e-01 |
Pos logit contributions | |
cess | +0.041 |
ck | +0.040 |
charged | +0.038 |
oxy | +0.037 |
tune | +0.037 |
Neg logit contributions | |
wrong | -0.019 |
sky | -0.019 |
done | -0.018 |
missing | -0.018 |
can | -0.018 |
Token | , |
Token ID | 11 |
Feature activation | +1.52e+00 |
Pos logit contributions | |
cess | +0.023 |
ck | +0.021 |
charged | +0.021 |
abbit | +0.020 |
aret | +0.020 |
Neg logit contributions | |
wrong | -0.012 |
done | -0.012 |
sky | -0.011 |
missing | -0.011 |
can | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +2.22e+00 |
Pos logit contributions | |
cess | +0.041 |
ck | +0.038 |
charged | +0.037 |
tune | +0.036 |
aret | +0.035 |
Neg logit contributions | |
wrong | -0.027 |
done | -0.025 |
sky | -0.025 |
missing | -0.024 |
can | -0.024 |
Token | when |
Token ID | 618 |
Feature activation | -1.76e+00 |
Pos logit contributions | |
cess | +0.058 |
ck | +0.054 |
tune | +0.051 |
charged | +0.051 |
aret | +0.050 |
Neg logit contributions | |
wrong | -0.041 |
done | -0.038 |
sky | -0.038 |
missing | -0.037 |
am | -0.036 |
Token | it |
Token ID | 340 |
Feature activation | -7.32e+00 |
Pos logit contributions | |
done | +0.030 |
wrong | +0.030 |
boards | +0.029 |
missing | +0.028 |
sky | +0.028 |
Neg logit contributions | |
cess | -0.044 |
ck | -0.041 |
charged | -0.040 |
tune | -0.039 |
feat | -0.039 |
Token | was |
Token ID | 373 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
wrong | +0.098 |
missing | +0.086 |
biggest | +0.086 |
sky | +0.079 |
living | +0.079 |
Neg logit contributions | |
ck | -0.202 |
cess | -0.189 |
tune | -0.183 |
feat | -0.181 |
eter | -0.173 |
Token | time |
Token ID | 640 |
Feature activation | -3.20e+00 |
Pos logit contributions | |
am | +0.050 |
boards | +0.050 |
can | +0.048 |
has | +0.046 |
wrong | +0.042 |
Neg logit contributions | |
ck | -0.080 |
aret | -0.080 |
cess | -0.078 |
ishment | -0.075 |
tune | -0.075 |
Token | to |
Token ID | 284 |
Feature activation | +1.64e+00 |
Pos logit contributions | |
am | +0.069 |
sky | +0.069 |
toilet | +0.068 |
coolest | +0.067 |
biggest | +0.065 |
Neg logit contributions | |
rily | -0.051 |
aret | -0.051 |
fully | -0.050 |
cess | -0.048 |
ck | -0.048 |
Token | leave |
Token ID | 2666 |
Feature activation | -4.84e-01 |
Pos logit contributions | |
cess | +0.048 |
ck | +0.044 |
charged | +0.043 |
aret | +0.043 |
oxy | +0.041 |
Neg logit contributions | |
wrong | -0.026 |
done | -0.025 |
sky | -0.025 |
missing | -0.024 |
made | -0.024 |
Token | , |
Token ID | 11 |
Feature activation | +7.13e-01 |
Pos logit contributions | |
done | +0.008 |
wrong | +0.008 |
sky | +0.008 |
missing | +0.007 |
has | +0.007 |
Neg logit contributions | |
cess | -0.013 |
charged | -0.012 |
ck | -0.012 |
aret | -0.012 |
ishment | -0.011 |
Token | of |
Token ID | 286 |
Feature activation | +1.44e+00 |
Pos logit contributions | |
wrong | +0.000 |
done | +0.000 |
made | +0.000 |
sky | +0.000 |
missing | +0.000 |
Neg logit contributions | |
cess | -0.001 |
charged | -0.001 |
ck | -0.001 |
aret | -0.001 |
abbit | -0.001 |
Token | fun |
Token ID | 1257 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
cess | +0.049 |
ck | +0.047 |
charged | +0.046 |
oxy | +0.045 |
aret | +0.045 |
Neg logit contributions | |
wrong | -0.014 |
done | -0.014 |
sky | -0.013 |
missing | -0.013 |
made | -0.013 |
Token | . |
Token ID | 13 |
Feature activation | -4.62e-02 |
Pos logit contributions | |
cess | +0.052 |
charged | +0.048 |
ck | +0.047 |
aret | +0.046 |
abbit | +0.046 |
Neg logit contributions | |
wrong | -0.029 |
done | -0.029 |
made | -0.028 |
sky | -0.028 |
missing | -0.027 |
Token | But |
Token ID | 887 |
Feature activation | -1.77e-02 |
Pos logit contributions | |
wrong | +0.001 |
done | +0.001 |
sky | +0.001 |
made | +0.001 |
missing | +0.001 |
Neg logit contributions | |
cess | -0.001 |
charged | -0.001 |
ck | -0.001 |
aret | -0.001 |
abbit | -0.001 |
Token | when |
Token ID | 618 |
Feature activation | -2.50e+00 |
Pos logit contributions | |
done | +0.000 |
wrong | +0.000 |
sky | +0.000 |
boards | +0.000 |
can | +0.000 |
Neg logit contributions | |
cess | -0.000 |
ck | -0.000 |
charged | -0.000 |
aret | -0.000 |
ishment | -0.000 |
Token | it |
Token ID | 340 |
Feature activation | -7.31e+00 |
Pos logit contributions | |
done | +0.041 |
wrong | +0.040 |
missing | +0.038 |
sky | +0.038 |
made | +0.037 |
Neg logit contributions | |
cess | -0.068 |
charged | -0.061 |
ck | -0.061 |
aret | -0.061 |
chick | -0.058 |
Token | was |
Token ID | 373 |
Feature activation | -3.38e+00 |
Pos logit contributions | |
wrong | +0.071 |
done | +0.061 |
best | +0.059 |
sky | +0.058 |
missing | +0.057 |
Neg logit contributions | |
cess | -0.236 |
ck | -0.228 |
tune | -0.213 |
charged | -0.212 |
oing | -0.210 |
Token | time |
Token ID | 640 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
can | +0.043 |
am | +0.041 |
has | +0.039 |
boards | +0.039 |
wrong | +0.039 |
Neg logit contributions | |
cess | -0.097 |
ck | -0.093 |
aret | -0.093 |
charged | -0.090 |
tune | -0.089 |
Token | to |
Token ID | 284 |
Feature activation | +1.80e+00 |
Pos logit contributions | |
sky | +0.076 |
coolest | +0.074 |
biggest | +0.074 |
loads | +0.072 |
boards | +0.071 |
Neg logit contributions | |
aret | -0.065 |
cess | -0.065 |
ck | -0.058 |
.'" | -0.057 |
heads | -0.057 |
Token | go |
Token ID | 467 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
cess | +0.060 |
charged | +0.056 |
abbit | +0.056 |
oxy | +0.054 |
ymes | +0.054 |
Neg logit contributions | |
done | -0.023 |
wrong | -0.021 |
made | -0.021 |
can | -0.020 |
sky | -0.020 |
Token | home |
Token ID | 1363 |
Feature activation | -5.16e-01 |
Pos logit contributions | |
done | +0.028 |
boards | +0.028 |
wrong | +0.028 |
sky | +0.028 |
made | +0.028 |
Neg logit contributions | |
cess | -0.033 |
ck | -0.028 |
charged | -0.028 |
aret | -0.028 |
feat | -0.027 |
Token | open |
Token ID | 1280 |
Feature activation | -2.19e-01 |
Pos logit contributions | |
cess | +0.057 |
ck | +0.051 |
charged | +0.050 |
aret | +0.049 |
abbit | +0.047 |
Neg logit contributions | |
wrong | -0.046 |
done | -0.044 |
sky | -0.044 |
missing | -0.042 |
made | -0.042 |
Token | the |
Token ID | 262 |
Feature activation | -3.82e+00 |
Pos logit contributions | |
done | +0.004 |
wrong | +0.004 |
had | +0.004 |
am | +0.004 |
has | +0.004 |
Neg logit contributions | |
cess | -0.005 |
rily | -0.005 |
ck | -0.005 |
aret | -0.005 |
ishment | -0.005 |
Token | jar |
Token ID | 17379 |
Feature activation | -4.87e+00 |
Pos logit contributions | |
ained | +0.079 |
done | +0.078 |
made | +0.072 |
Everything | +0.070 |
arrived | +0.070 |
Neg logit contributions | |
ymes | -0.069 |
wrist | -0.068 |
wrists | -0.066 |
cess | -0.064 |
chick | -0.063 |
Token | and |
Token ID | 290 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
wrong | +0.085 |
done | +0.083 |
missing | +0.079 |
am | +0.078 |
has | +0.076 |
Neg logit contributions | |
cess | -0.115 |
rily | -0.113 |
eter | -0.105 |
ress | -0.102 |
aret | -0.102 |
Token | when |
Token ID | 618 |
Feature activation | -1.92e+00 |
Pos logit contributions | |
cess | +0.062 |
ck | +0.057 |
tune | +0.052 |
ishment | +0.052 |
eter | +0.052 |
Neg logit contributions | |
wrong | -0.048 |
sky | -0.045 |
done | -0.044 |
best | -0.043 |
missing | -0.043 |
Token | it |
Token ID | 340 |
Feature activation | -7.31e+00 |
Pos logit contributions | |
done | +0.022 |
wrong | +0.022 |
sky | +0.019 |
missing | +0.019 |
made | +0.019 |
Neg logit contributions | |
cess | -0.062 |
charged | -0.056 |
aret | -0.055 |
ck | -0.055 |
abbit | -0.054 |
Token | finally |
Token ID | 3443 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
wrong | +0.103 |
sky | +0.086 |
best | +0.085 |
living | +0.082 |
missing | +0.081 |
Neg logit contributions | |
cess | -0.197 |
ck | -0.195 |
eter | -0.180 |
rily | -0.179 |
owed | -0.178 |
Token | opened |
Token ID | 4721 |
Feature activation | -1.14e+00 |
Pos logit contributions | |
wrong | +0.080 |
sky | +0.075 |
best | +0.074 |
biggest | +0.073 |
most | +0.069 |
Neg logit contributions | |
ck | -0.112 |
cess | -0.104 |
rily | -0.095 |
ided | -0.095 |
feat | -0.093 |
Token | , |
Token ID | 11 |
Feature activation | -6.29e-01 |
Pos logit contributions | |
done | +0.021 |
wrong | +0.020 |
am | +0.019 |
has | +0.018 |
sky | +0.018 |
Neg logit contributions | |
cess | -0.029 |
rily | -0.029 |
ck | -0.027 |
aret | -0.027 |
ishment | -0.026 |
Token | the |
Token ID | 262 |
Feature activation | -2.79e+00 |
Pos logit contributions | |
done | +0.009 |
wrong | +0.009 |
missing | +0.008 |
sky | +0.008 |
am | +0.008 |
Neg logit contributions | |
cess | -0.019 |
rily | -0.017 |
ck | -0.017 |
aret | -0.016 |
ishment | -0.016 |
Token | playground |
Token ID | 24817 |
Feature activation | -2.23e+00 |
Pos logit contributions | |
ained | +0.053 |
done | +0.052 |
rob | +0.050 |
arrived | +0.050 |
Everything | +0.050 |
Neg logit contributions | |
wrist | -0.053 |
ymes | -0.050 |
wrists | -0.050 |
cess | -0.049 |
chick | -0.049 |
Token | time |
Token ID | 640 |
Feature activation | +3.64e-01 |
Pos logit contributions | |
wrong | +0.001 |
done | +0.001 |
missing | +0.001 |
can | +0.001 |
sky | +0.001 |
Neg logit contributions | |
cess | -0.003 |
ck | -0.003 |
tune | -0.003 |
aret | -0.003 |
abbit | -0.003 |
Token | playing |
Token ID | 2712 |
Feature activation | -4.98e-01 |
Pos logit contributions | |
cess | +0.010 |
ck | +0.009 |
charged | +0.009 |
feat | +0.009 |
oing | +0.009 |
Neg logit contributions | |
wrong | -0.006 |
sky | -0.006 |
done | -0.006 |
missing | -0.006 |
can | -0.006 |
Token | , |
Token ID | 11 |
Feature activation | +1.85e+00 |
Pos logit contributions | |
done | +0.009 |
wrong | +0.009 |
sky | +0.008 |
missing | +0.008 |
made | +0.008 |
Neg logit contributions | |
cess | -0.014 |
charged | -0.012 |
ck | -0.012 |
abbit | -0.012 |
aret | -0.012 |
Token | and |
Token ID | 290 |
Feature activation | +2.15e+00 |
Pos logit contributions | |
cess | +0.051 |
ck | +0.046 |
charged | +0.045 |
tune | +0.045 |
eter | +0.044 |
Neg logit contributions | |
wrong | -0.032 |
done | -0.031 |
sky | -0.029 |
missing | -0.029 |
can | -0.028 |
Token | when |
Token ID | 618 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
cess | +0.057 |
ck | +0.052 |
tune | +0.049 |
eter | +0.048 |
charged | +0.048 |
Neg logit contributions | |
wrong | -0.040 |
sky | -0.037 |
missing | -0.036 |
boards | -0.035 |
can | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | -7.29e+00 |
Pos logit contributions | |
wrong | +0.036 |
boards | +0.035 |
done | +0.035 |
sky | +0.034 |
missing | +0.034 |
Neg logit contributions | |
cess | -0.050 |
ck | -0.046 |
charged | -0.044 |
feat | -0.043 |
eter | -0.043 |
Token | was |
Token ID | 373 |
Feature activation | -3.17e+00 |
Pos logit contributions | |
wrong | +0.099 |
missing | +0.089 |
biggest | +0.087 |
best | +0.083 |
seeker | +0.081 |
Neg logit contributions | |
ck | -0.190 |
cess | -0.190 |
feat | -0.180 |
ie | -0.176 |
tune | -0.175 |
Token | time |
Token ID | 640 |
Feature activation | -3.45e+00 |
Pos logit contributions | |
am | +0.048 |
can | +0.048 |
boards | +0.047 |
has | +0.044 |
wrong | +0.041 |
Neg logit contributions | |
aret | -0.082 |
ck | -0.080 |
cess | -0.080 |
tune | -0.076 |
ishment | -0.074 |
Token | to |
Token ID | 284 |
Feature activation | +1.45e+00 |
Pos logit contributions | |
sky | +0.073 |
coolest | +0.072 |
am | +0.071 |
toilet | +0.070 |
biggest | +0.069 |
Neg logit contributions | |
aret | -0.059 |
rily | -0.057 |
cess | -0.054 |
fully | -0.053 |
.'" | -0.052 |
Token | go |
Token ID | 467 |
Feature activation | -1.55e+00 |
Pos logit contributions | |
cess | +0.042 |
ck | +0.039 |
charged | +0.038 |
aret | +0.038 |
oxy | +0.037 |
Neg logit contributions | |
wrong | -0.023 |
done | -0.022 |
sky | -0.022 |
missing | -0.022 |
made | -0.021 |
Token | home |
Token ID | 1363 |
Feature activation | -3.15e-01 |
Pos logit contributions | |
boards | +0.034 |
wrong | +0.034 |
made | +0.033 |
done | +0.033 |
sky | +0.032 |
Neg logit contributions | |
cess | -0.027 |
ck | -0.024 |
aret | -0.024 |
feat | -0.023 |
charged | -0.023 |
Token | always |
Token ID | 1464 |
Feature activation | +2.73e+00 |
Pos logit contributions | |
cess | +0.082 |
ck | +0.079 |
aret | +0.075 |
charged | +0.074 |
abbit | +0.071 |
Neg logit contributions | |
boards | -0.059 |
sky | -0.056 |
wrong | -0.056 |
missing | -0.054 |
done | -0.052 |
Token | think |
Token ID | 892 |
Feature activation | -1.20e+00 |
Pos logit contributions | |
cess | +0.058 |
ck | +0.052 |
charged | +0.050 |
tune | +0.048 |
aret | +0.048 |
Neg logit contributions | |
boards | -0.065 |
sky | -0.064 |
wrong | -0.061 |
missing | -0.060 |
done | -0.058 |
Token | about |
Token ID | 546 |
Feature activation | +1.57e+00 |
Pos logit contributions | |
done | +0.018 |
wrong | +0.018 |
sky | +0.017 |
missing | +0.017 |
made | +0.017 |
Neg logit contributions | |
cess | -0.035 |
ck | -0.032 |
aret | -0.031 |
charged | -0.031 |
abbit | -0.031 |
Token | our |
Token ID | 674 |
Feature activation | +2.60e-01 |
Pos logit contributions | |
cess | +0.042 |
charged | +0.038 |
ck | +0.037 |
abbit | +0.036 |
aret | +0.036 |
Neg logit contributions | |
done | -0.029 |
wrong | -0.028 |
sky | -0.028 |
made | -0.028 |
missing | -0.027 |
Token | safety |
Token ID | 3747 |
Feature activation | -5.10e-03 |
Pos logit contributions | |
cess | +0.007 |
ck | +0.006 |
charged | +0.006 |
aret | +0.006 |
oxy | +0.006 |
Neg logit contributions | |
wrong | -0.005 |
done | -0.005 |
sky | -0.005 |
can | -0.005 |
made | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
done | +0.000 |
wrong | +0.000 |
missing | +0.000 |
made | +0.000 |
sky | +0.000 |
Neg logit contributions | |
cess | -0.000 |
charged | -0.000 |
abbit | -0.000 |
aret | -0.000 |
ck | -0.000 |
Token | be |
Token ID | 307 |
Feature activation | +6.32e-01 |
Pos logit contributions | |
cess | +0.111 |
charged | +0.100 |
ck | +0.099 |
tune | +0.096 |
aret | +0.096 |
Neg logit contributions | |
sky | -0.098 |
boards | -0.098 |
wrong | -0.096 |
done | -0.094 |
missing | -0.093 |
Token | mindful |
Token ID | 33798 |
Feature activation | +5.60e-01 |
Pos logit contributions | |
cess | +0.014 |
ck | +0.012 |
charged | +0.012 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
sky | -0.014 |
wrong | -0.014 |
done | -0.014 |
can | -0.014 |
missing | -0.014 |
Token | of |
Token ID | 286 |
Feature activation | +2.26e+00 |
Pos logit contributions | |
cess | +0.015 |
charged | +0.014 |
aret | +0.014 |
ck | +0.014 |
abbit | +0.014 |
Neg logit contributions | |
done | -0.009 |
missing | -0.009 |
sky | -0.009 |
wrong | -0.009 |
made | -0.008 |
Token | what |
Token ID | 644 |
Feature activation | -2.89e+00 |
Pos logit contributions | |
cess | +0.058 |
charged | +0.052 |
ck | +0.052 |
aret | +0.051 |
abbit | +0.050 |
Neg logit contributions | |
done | -0.044 |
sky | -0.043 |
wrong | -0.043 |
made | -0.042 |
missing | -0.042 |
Token | we |
Token ID | 356 |
Feature activation | -1.93e+00 |
Pos logit contributions | |
done | +0.047 |
sky | +0.046 |
missing | +0.045 |
wrong | +0.045 |
made | +0.042 |
Neg logit contributions | |
cess | -0.078 |
ck | -0.071 |
charged | -0.070 |
abbit | -0.069 |
aret | -0.069 |
Token | and |
Token ID | 290 |
Feature activation | +3.13e+00 |
Pos logit contributions | |
wrong | +0.009 |
done | +0.009 |
missing | +0.009 |
sky | +0.009 |
had | +0.008 |
Neg logit contributions | |
cess | -0.013 |
charged | -0.012 |
eter | -0.011 |
rily | -0.011 |
.'" | -0.011 |
Token | I |
Token ID | 314 |
Feature activation | -9.77e-01 |
Pos logit contributions | |
cess | +0.081 |
charged | +0.069 |
ck | +0.068 |
tune | +0.066 |
aret | +0.066 |
Neg logit contributions | |
wrong | -0.056 |
done | -0.055 |
sky | -0.054 |
boards | -0.054 |
missing | -0.053 |
Token | saw |
Token ID | 2497 |
Feature activation | -1.13e+00 |
Pos logit contributions | |
wrong | +0.016 |
sky | +0.015 |
boards | +0.014 |
best | +0.014 |
missing | +0.014 |
Neg logit contributions | |
ck | -0.027 |
cess | -0.026 |
charged | -0.025 |
eter | -0.024 |
feat | -0.024 |
Token | something |
Token ID | 1223 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
done | +0.017 |
wrong | +0.017 |
missing | +0.016 |
had | +0.016 |
sky | +0.016 |
Neg logit contributions | |
cess | -0.033 |
charged | -0.030 |
ck | -0.030 |
aret | -0.030 |
abbit | -0.029 |
Token | go |
Token ID | 467 |
Feature activation | +4.25e-01 |
Pos logit contributions | |
done | +0.035 |
sky | +0.034 |
Everything | +0.030 |
| +0.030 |
boards | +0.030 |
Neg logit contributions | |
ck | -0.055 |
cess | -0.055 |
keys | -0.054 |
rily | -0.053 |
chick | -0.051 |
Token | po |
Token ID | 745 |
Feature activation | +4.78e+00 |
Pos logit contributions | |
cess | +0.012 |
ck | +0.011 |
aret | +0.011 |
charged | +0.011 |
rily | +0.010 |
Neg logit contributions | |
sky | -0.006 |
done | -0.006 |
best | -0.006 |
boards | -0.006 |
made | -0.005 |
Token | of |
Token ID | 1659 |
Feature activation | +7.32e-01 |
Pos logit contributions | |
cess | +0.097 |
ck | +0.085 |
aret | +0.085 |
charged | +0.085 |
abbit | +0.081 |
Neg logit contributions | |
wrong | -0.116 |
done | -0.115 |
made | -0.113 |
sky | -0.112 |
can | -0.111 |
Token | ! |
Token ID | 0 |
Feature activation | -1.80e-01 |
Pos logit contributions | |
cess | +0.019 |
charged | +0.017 |
ck | +0.017 |
aret | +0.016 |
abbit | +0.016 |
Neg logit contributions | |
wrong | -0.015 |
done | -0.015 |
made | -0.014 |
sky | -0.014 |
can | -0.014 |
Token | Then |
Token ID | 3244 |
Feature activation | -4.13e-01 |
Pos logit contributions | |
wrong | +0.003 |
done | +0.003 |
made | +0.003 |
sky | +0.003 |
missing | +0.003 |
Neg logit contributions | |
cess | -0.005 |
charged | -0.005 |
ck | -0.005 |
aret | -0.005 |
abbit | -0.005 |
Token | the |
Token ID | 262 |
Feature activation | -1.39e+00 |
Pos logit contributions | |
wrong | +0.008 |
done | +0.008 |
missing | +0.008 |
sky | +0.007 |
made | +0.007 |
Neg logit contributions | |
cess | -0.011 |
ck | -0.009 |
charged | -0.009 |
aret | -0.009 |
abbit | -0.009 |
Token | pen |
Token ID | 3112 |
Feature activation | -4.05e+00 |
Pos logit contributions | |
done | +0.028 |
am | +0.026 |
made | +0.026 |
has | +0.025 |
wrong | +0.025 |
Neg logit contributions | |
cess | -0.030 |
tune | -0.030 |
chick | -0.029 |
feat | -0.028 |
aret | -0.027 |
Token | I |
Token ID | 314 |
Feature activation | -1.04e-01 |
Pos logit contributions | |
wrong | +0.002 |
done | +0.002 |
made | +0.002 |
can | +0.002 |
sky | +0.002 |
Neg logit contributions | |
cess | -0.004 |
charged | -0.004 |
ck | -0.004 |
aret | -0.004 |
abbit | -0.004 |
Token | have |
Token ID | 423 |
Feature activation | -6.25e-01 |
Pos logit contributions | |
wrong | +0.002 |
sky | +0.002 |
done | +0.002 |
missing | +0.002 |
boards | +0.002 |
Neg logit contributions | |
cess | -0.003 |
ck | -0.002 |
oxy | -0.002 |
charged | -0.002 |
abbit | -0.002 |
Token | some |
Token ID | 617 |
Feature activation | -7.88e-02 |
Pos logit contributions | |
sky | +0.009 |
wrong | +0.009 |
done | +0.009 |
missing | +0.008 |
has | +0.008 |
Neg logit contributions | |
ck | -0.017 |
cess | -0.017 |
oxy | -0.017 |
chick | -0.016 |
charged | -0.016 |
Token | coffee |
Token ID | 6891 |
Feature activation | -6.18e-01 |
Pos logit contributions | |
sky | +0.001 |
wrong | +0.001 |
missing | +0.001 |
had | +0.001 |
done | +0.001 |
Neg logit contributions | |
cess | -0.002 |
ck | -0.002 |
oxy | -0.002 |
charged | -0.002 |
abbit | -0.002 |
Token | , |
Token ID | 11 |
Feature activation | +3.66e+00 |
Pos logit contributions | |
sky | +0.011 |
missing | +0.010 |
wrong | +0.010 |
done | +0.010 |
had | +0.010 |
Neg logit contributions | |
cess | -0.015 |
rily | -0.014 |
.'" | -0.014 |
eter | -0.013 |
oxy | -0.013 |
Token | mom |
Token ID | 1995 |
Feature activation | +5.84e+00 |
Pos logit contributions | |
cess | +0.089 |
charged | +0.082 |
ck | +0.079 |
abbit | +0.077 |
aret | +0.077 |
Neg logit contributions | |
wrong | -0.078 |
made | -0.078 |
can | -0.077 |
done | -0.077 |
sky | -0.075 |
Token | my |
Token ID | 1820 |
Feature activation | +3.23e+00 |
Pos logit contributions | |
cess | +0.174 |
charged | +0.159 |
ck | +0.156 |
abbit | +0.153 |
aret | +0.153 |
Neg logit contributions | |
done | -0.093 |
wrong | -0.092 |
made | -0.090 |
can | -0.087 |
sky | -0.087 |
Token | ?" |
Token ID | 1701 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
cess | +0.101 |
charged | +0.095 |
ck | +0.092 |
abbit | +0.091 |
oxy | +0.091 |
Neg logit contributions | |
made | -0.048 |
wrong | -0.048 |
done | -0.047 |
can | -0.045 |
had | -0.044 |
Token | she |
Token ID | 673 |
Feature activation | +1.41e+00 |
Pos logit contributions | |
cess | +0.082 |
charged | +0.081 |
eter | +0.077 |
ymes | +0.076 |
abbit | +0.075 |
Neg logit contributions | |
made | -0.029 |
had | -0.028 |
on | -0.026 |
wrong | -0.025 |
sky | -0.025 |
Token | asked |
Token ID | 1965 |
Feature activation | +5.42e-01 |
Pos logit contributions | |
charged | +0.048 |
cess | +0.048 |
eter | +0.047 |
oing | +0.047 |
ymes | +0.045 |
Neg logit contributions | |
made | -0.020 |
had | -0.020 |
done | -0.017 |
wrong | -0.017 |
did | -0.017 |
Token | . |
Token ID | 13 |
Feature activation | +9.46e-01 |
Pos logit contributions | |
cess | +0.021 |
oing | +0.019 |
charged | +0.019 |
abbit | +0.019 |
keys | +0.019 |
Neg logit contributions | |
on | -0.007 |
all | -0.006 |
wrong | -0.006 |
made | -0.005 |
had | -0.005 |
Token | . |
Token ID | 13 |
Feature activation | -6.75e-01 |
Pos logit contributions | |
cess | +0.107 |
charged | +0.098 |
ck | +0.096 |
aret | +0.095 |
oxy | +0.094 |
Neg logit contributions | |
made | -0.047 |
had | -0.046 |
wrong | -0.046 |
done | -0.045 |
can | -0.044 |
Token | She |
Token ID | 1375 |
Feature activation | -8.97e-01 |
Pos logit contributions | |
wrong | +0.010 |
done | +0.010 |
sky | +0.009 |
made | +0.009 |
can | +0.009 |
Neg logit contributions | |
cess | -0.021 |
ck | -0.020 |
charged | -0.019 |
aret | -0.019 |
oxy | -0.019 |
Token | loved |
Token ID | 6151 |
Feature activation | +3.74e-01 |
Pos logit contributions | |
wrong | +0.014 |
done | +0.013 |
sky | +0.013 |
best | +0.013 |
missing | +0.013 |
Neg logit contributions | |
cess | -0.024 |
ck | -0.023 |
oxy | -0.023 |
charged | -0.022 |
abbit | -0.022 |
Token | to |
Token ID | 284 |
Feature activation | +2.76e+00 |
Pos logit contributions | |
cess | +0.013 |
charged | +0.012 |
ck | +0.012 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
done | -0.004 |
wrong | -0.004 |
made | -0.004 |
sky | -0.004 |
missing | -0.004 |
Token | paint |
Token ID | 7521 |
Feature activation | +6.01e-01 |
Pos logit contributions | |
cess | +0.084 |
charged | +0.078 |
abbit | +0.075 |
aret | +0.075 |
ck | +0.075 |
Neg logit contributions | |
done | -0.043 |
made | -0.043 |
can | -0.041 |
had | -0.041 |
wrong | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +4.77e+00 |
Pos logit contributions | |
cess | +0.013 |
eter | +0.012 |
ck | +0.012 |
ymes | +0.012 |
wrists | +0.012 |
Neg logit contributions | |
done | -0.012 |
wrong | -0.011 |
missing | -0.011 |
arrived | -0.011 |
had | -0.010 |
Token | make |
Token ID | 787 |
Feature activation | +3.16e-01 |
Pos logit contributions | |
ck | +0.098 |
cess | +0.098 |
tune | +0.097 |
wrists | +0.095 |
aret | +0.094 |
Neg logit contributions | |
wrong | -0.103 |
done | -0.095 |
missing | -0.093 |
| -0.091 |
boards | -0.089 |
Token | things |
Token ID | 1243 |
Feature activation | +1.95e+00 |
Pos logit contributions | |
cess | +0.010 |
charged | +0.009 |
ck | +0.009 |
aret | +0.009 |
abbit | +0.008 |
Neg logit contributions | |
wrong | -0.005 |
done | -0.005 |
made | -0.004 |
sky | -0.004 |
missing | -0.004 |
Token | bright |
Token ID | 6016 |
Feature activation | +3.09e+00 |
Pos logit contributions | |
ck | +0.049 |
cess | +0.049 |
rily | +0.045 |
eter | +0.045 |
oing | +0.044 |
Neg logit contributions | |
done | -0.030 |
sky | -0.030 |
arrived | -0.030 |
wrong | -0.030 |
| -0.029 |
Token | and |
Token ID | 290 |
Feature activation | +4.16e+00 |
Pos logit contributions | |
cess | +0.069 |
eter | +0.065 |
rily | +0.063 |
ck | +0.061 |
feat | +0.060 |
Neg logit contributions | |
done | -0.066 |
wrong | -0.065 |
missing | -0.061 |
had | -0.058 |
arrived | -0.058 |
Token | full |
Token ID | 1336 |
Feature activation | -2.02e+00 |
Pos logit contributions | |
cess | +0.100 |
ck | +0.097 |
tune | +0.094 |
aret | +0.092 |
wrists | +0.092 |
Neg logit contributions | |
wrong | -0.074 |
| -0.068 |
done | -0.068 |
missing | -0.068 |
sky | -0.067 |
Token | ma |
Token ID | 2611 |
Feature activation | +3.69e+00 |
Pos logit contributions | |
cess | +0.122 |
charged | +0.106 |
abbit | +0.104 |
ck | +0.104 |
oxy | +0.100 |
Neg logit contributions | |
done | -0.135 |
wrong | -0.134 |
made | -0.128 |
missing | -0.125 |
had | -0.125 |
Token | , |
Token ID | 11 |
Feature activation | +5.61e-01 |
Pos logit contributions | |
cess | +0.104 |
aret | +0.093 |
charged | +0.092 |
abbit | +0.091 |
ck | +0.091 |
Neg logit contributions | |
done | -0.062 |
sky | -0.060 |
wrong | -0.059 |
made | -0.057 |
missing | -0.056 |
Token | � |
Token ID | 6184 |
Feature activation | +3.40e+00 |
Pos logit contributions | |
cess | +0.018 |
charged | +0.017 |
ck | +0.017 |
aret | +0.016 |
abbit | +0.016 |
Neg logit contributions | |
done | -0.007 |
wrong | -0.007 |
sky | -0.007 |
made | -0.007 |
missing | -0.006 |
Token | � |
Token ID | 95 |
Feature activation | +3.33e+00 |
Pos logit contributions | |
cess | +0.082 |
aret | +0.078 |
charged | +0.078 |
ck | +0.077 |
abbit | +0.076 |
Neg logit contributions | |
wrong | -0.069 |
done | -0.065 |
missing | -0.064 |
can | -0.063 |
made | -0.062 |
Token | € |
Token ID | 26391 |
Feature activation | +2.04e+00 |
Pos logit contributions | |
cess | +0.083 |
charged | +0.078 |
ck | +0.077 |
aret | +0.075 |
abbit | +0.074 |
Neg logit contributions | |
wrong | -0.067 |
done | -0.066 |
made | -0.063 |
sky | -0.063 |
missing | -0.062 |
Token | � |
Token ID | 129 |
Feature activation | +4.43e+00 |
Pos logit contributions | |
cess | +0.050 |
charged | +0.046 |
ck | +0.045 |
aret | +0.044 |
abbit | +0.044 |
Neg logit contributions | |
done | -0.042 |
wrong | -0.041 |
made | -0.040 |
sky | -0.040 |
can | -0.039 |
Token | � |
Token ID | 241 |
Feature activation | +2.06e+00 |
Pos logit contributions | |
cess | +0.109 |
charged | +0.102 |
ymes | +0.100 |
abbit | +0.099 |
chick | +0.099 |
Neg logit contributions | |
wrong | -0.091 |
done | -0.089 |
missing | -0.087 |
made | -0.083 |
sky | -0.081 |
Token | Can |
Token ID | 6090 |
Feature activation | -2.74e-01 |
Pos logit contributions | |
cess | +0.055 |
oxy | +0.050 |
ck | +0.049 |
charged | +0.048 |
abbit | +0.048 |
Neg logit contributions | |
done | -0.038 |
wrong | -0.037 |
missing | -0.035 |
sky | -0.034 |
best | -0.034 |
Token | I |
Token ID | 314 |
Feature activation | +5.33e-01 |
Pos logit contributions | |
done | +0.005 |
wrong | +0.005 |
sky | +0.005 |
made | +0.005 |
missing | +0.005 |
Neg logit contributions | |
cess | -0.007 |
charged | -0.006 |
ck | -0.006 |
aret | -0.006 |
oxy | -0.006 |
Token | fix |
Token ID | 4259 |
Feature activation | +1.78e+00 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.012 |
ck | +0.012 |
aret | +0.012 |
abbit | +0.012 |
Neg logit contributions | |
wrong | -0.010 |
done | -0.010 |
sky | -0.010 |
missing | -0.009 |
made | -0.009 |
Token | the |
Token ID | 262 |
Feature activation | -4.61e-01 |
Pos logit contributions | |
cess | +0.047 |
ck | +0.042 |
charged | +0.042 |
aret | +0.042 |
oxy | +0.041 |
Neg logit contributions | |
done | -0.033 |
wrong | -0.032 |
sky | -0.032 |
missing | -0.031 |
made | -0.031 |
Token | made |
Token ID | 925 |
Feature activation | -1.35e+00 |
Pos logit contributions | |
cess | +0.013 |
ck | +0.012 |
aret | +0.012 |
oxy | +0.011 |
chick | +0.011 |
Neg logit contributions | |
wrong | -0.009 |
sky | -0.008 |
missing | -0.008 |
done | -0.008 |
best | -0.008 |
Token | him |
Token ID | 683 |
Feature activation | +1.90e+00 |
Pos logit contributions | |
wrong | +0.016 |
done | +0.015 |
made | +0.014 |
sky | +0.014 |
missing | +0.014 |
Neg logit contributions | |
cess | -0.046 |
charged | -0.042 |
ck | -0.042 |
aret | -0.041 |
abbit | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | +5.17e-01 |
Pos logit contributions | |
cess | +0.070 |
charged | +0.065 |
aret | +0.064 |
abbit | +0.063 |
ck | +0.063 |
Neg logit contributions | |
done | -0.018 |
wrong | -0.018 |
made | -0.017 |
missing | -0.016 |
on | -0.015 |
Token | big |
Token ID | 1263 |
Feature activation | +2.14e+00 |
Pos logit contributions | |
cess | +0.016 |
charged | +0.014 |
ck | +0.014 |
abbit | +0.014 |
aret | +0.014 |
Neg logit contributions | |
done | -0.007 |
wrong | -0.007 |
made | -0.007 |
sky | -0.007 |
had | -0.007 |
Token | sandwich |
Token ID | 20433 |
Feature activation | -6.64e-01 |
Pos logit contributions | |
cess | +0.058 |
charged | +0.053 |
ck | +0.052 |
aret | +0.051 |
abbit | +0.051 |
Neg logit contributions | |
wrong | -0.039 |
done | -0.039 |
made | -0.037 |
sky | -0.037 |
missing | -0.036 |
Token | for |
Token ID | 329 |
Feature activation | +9.11e-01 |
Pos logit contributions | |
done | +0.011 |
wrong | +0.011 |
sky | +0.010 |
missing | +0.010 |
can | +0.010 |
Neg logit contributions | |
cess | -0.018 |
ck | -0.016 |
aret | -0.016 |
rily | -0.016 |
charged | -0.016 |
Token | lunch |
Token ID | 9965 |
Feature activation | -1.71e+00 |
Pos logit contributions | |
cess | +0.028 |
charged | +0.026 |
aret | +0.025 |
ck | +0.025 |
abbit | +0.025 |
Neg logit contributions | |
done | -0.014 |
wrong | -0.014 |
made | -0.013 |
missing | -0.013 |
sky | -0.012 |
Token | . |
Token ID | 13 |
Feature activation | -4.06e-01 |
Pos logit contributions | |
wrong | +0.031 |
done | +0.030 |
sky | +0.030 |
can | +0.029 |
missing | +0.029 |
Neg logit contributions | |
cess | -0.044 |
charged | -0.039 |
ck | -0.039 |
aret | -0.039 |
oxy | -0.039 |
Token | Tim |
Token ID | 5045 |
Feature activation | +8.20e-01 |
Pos logit contributions | |
wrong | +0.007 |
done | +0.007 |
made | +0.007 |
sky | +0.007 |
missing | +0.007 |
Neg logit contributions | |
cess | -0.011 |
charged | -0.010 |
ck | -0.010 |
aret | -0.010 |
abbit | -0.010 |
Token | was |
Token ID | 373 |
Feature activation | -1.50e+00 |
Pos logit contributions | |
cess | +0.031 |
charged | +0.029 |
aret | +0.028 |
ck | +0.028 |
abbit | +0.028 |
Neg logit contributions | |
done | -0.007 |
made | -0.007 |
had | -0.007 |
wrong | -0.006 |
can | -0.006 |
Token | very |
Token ID | 845 |
Feature activation | +3.56e-01 |
Pos logit contributions | |
wrong | +0.016 |
done | +0.016 |
sky | +0.015 |
made | +0.014 |
can | +0.014 |
Neg logit contributions | |
cess | -0.052 |
charged | -0.048 |
ck | -0.047 |
aret | -0.047 |
abbit | -0.046 |
Token |
|
Token ID | 198 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
done | +0.002 |
wrong | +0.001 |
had | +0.001 |
made | +0.001 |
on | +0.001 |
Neg logit contributions | |
cess | -0.009 |
charged | -0.009 |
ishment | -0.009 |
eter | -0.008 |
aret | -0.008 |
Token | Tim |
Token ID | 14967 |
Feature activation | +2.09e-01 |
Pos logit contributions | |
cess | +0.029 |
aret | +0.026 |
charged | +0.026 |
rily | +0.026 |
ishment | +0.026 |
Neg logit contributions | |
made | -0.023 |
had | -0.022 |
done | -0.022 |
wrong | -0.021 |
on | -0.021 |
Token | did |
Token ID | 750 |
Feature activation | -1.73e+00 |
Pos logit contributions | |
cess | +0.007 |
charged | +0.007 |
ck | +0.006 |
aret | +0.006 |
abbit | +0.006 |
Neg logit contributions | |
made | -0.003 |
done | -0.003 |
wrong | -0.002 |
had | -0.002 |
can | -0.002 |
Token | not |
Token ID | 407 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
wrong | +0.022 |
done | +0.021 |
made | +0.020 |
missing | +0.020 |
had | +0.019 |
Neg logit contributions | |
cess | -0.058 |
charged | -0.053 |
aret | -0.052 |
abbit | -0.052 |
ck | -0.052 |
Token | want |
Token ID | 765 |
Feature activation | -1.95e+00 |
Pos logit contributions | |
cess | +0.031 |
ck | +0.028 |
charged | +0.028 |
aret | +0.027 |
oxy | +0.027 |
Neg logit contributions | |
wrong | -0.026 |
done | -0.026 |
sky | -0.026 |
missing | -0.025 |
made | -0.025 |
Token | to |
Token ID | 284 |
Feature activation | +3.08e+00 |
Pos logit contributions | |
wrong | +0.025 |
done | +0.025 |
made | +0.023 |
sky | +0.023 |
missing | +0.023 |
Neg logit contributions | |
cess | -0.063 |
charged | -0.058 |
ck | -0.058 |
aret | -0.057 |
abbit | -0.057 |
Token | share |
Token ID | 2648 |
Feature activation | +5.72e-01 |
Pos logit contributions | |
cess | +0.086 |
abbit | +0.080 |
rily | +0.080 |
charged | +0.079 |
aret | +0.079 |
Neg logit contributions | |
done | -0.057 |
made | -0.056 |
wrong | -0.056 |
all | -0.053 |
had | -0.052 |
Token | , |
Token ID | 11 |
Feature activation | +2.36e+00 |
Pos logit contributions | |
cess | +0.017 |
charged | +0.016 |
ck | +0.015 |
aret | +0.015 |
abbit | +0.015 |
Neg logit contributions | |
wrong | -0.009 |
done | -0.009 |
made | -0.009 |
sky | -0.008 |
missing | -0.008 |
Token | so |
Token ID | 523 |
Feature activation | +7.80e-01 |
Pos logit contributions | |
cess | +0.076 |
charged | +0.070 |
ck | +0.069 |
aret | +0.069 |
abbit | +0.068 |
Neg logit contributions | |
wrong | -0.030 |
done | -0.030 |
made | -0.028 |
sky | -0.028 |
missing | -0.028 |
Token | he |
Token ID | 339 |
Feature activation | -1.61e+00 |
Pos logit contributions | |
cess | +0.026 |
ck | +0.024 |
charged | +0.024 |
aret | +0.023 |
abbit | +0.023 |
Neg logit contributions | |
wrong | -0.009 |
done | -0.009 |
sky | -0.008 |
missing | -0.008 |
can | -0.008 |
Token | tried |
Token ID | 3088 |
Feature activation | -1.24e+00 |
Pos logit contributions | |
made | +0.023 |
done | +0.022 |
had | +0.022 |
wrong | +0.022 |
all | +0.020 |
Neg logit contributions | |
cess | -0.052 |
charged | -0.049 |
oing | -0.047 |
abbit | -0.047 |
aret | -0.047 |
Token | tree |
Token ID | 5509 |
Feature activation | +9.77e-01 |
Pos logit contributions | |
done | +0.001 |
wrong | +0.001 |
made | +0.001 |
had | +0.001 |
missing | +0.001 |
Neg logit contributions | |
cess | -0.002 |
tune | -0.002 |
chick | -0.002 |
aret | -0.002 |
charged | -0.002 |
Token | . |
Token ID | 13 |
Feature activation | +3.46e-01 |
Pos logit contributions | |
cess | +0.031 |
charged | +0.028 |
aret | +0.027 |
ck | +0.027 |
abbit | +0.027 |
Neg logit contributions | |
wrong | -0.015 |
done | -0.014 |
made | -0.014 |
missing | -0.014 |
on | -0.013 |
Token | She |
Token ID | 1375 |
Feature activation | -9.39e-01 |
Pos logit contributions | |
cess | +0.012 |
ck | +0.012 |
charged | +0.012 |
aret | +0.011 |
ymes | +0.011 |
Neg logit contributions | |
sky | -0.004 |
can | -0.004 |
made | -0.003 |
had | -0.003 |
wrong | -0.003 |
Token | asked |
Token ID | 1965 |
Feature activation | +2.08e-01 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.010 |
made | +0.010 |
had | +0.009 |
sky | +0.009 |
Neg logit contributions | |
cess | -0.033 |
charged | -0.030 |
ck | -0.030 |
aret | -0.029 |
abbit | -0.029 |
Token | her |
Token ID | 607 |
Feature activation | +1.03e+00 |
Pos logit contributions | |
cess | +0.007 |
charged | +0.007 |
oing | +0.007 |
ishment | +0.006 |
abbit | +0.006 |
Neg logit contributions | |
all | -0.003 |
made | -0.003 |
on | -0.003 |
done | -0.003 |
wrong | -0.003 |
Token | mom |
Token ID | 1995 |
Feature activation | +3.78e+00 |
Pos logit contributions | |
cess | +0.028 |
charged | +0.027 |
ck | +0.026 |
ishment | +0.026 |
oing | +0.025 |
Neg logit contributions | |
best | -0.020 |
can | -0.020 |
all | -0.020 |
made | -0.020 |
on | -0.020 |
Token | for |
Token ID | 329 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
cess | +0.073 |
aret | +0.068 |
God | +0.066 |
abbit | +0.064 |
Him | +0.063 |
Neg logit contributions | |
sky | -0.082 |
done | -0.082 |
wrong | -0.080 |
missing | -0.078 |
toilet | -0.075 |
Token | help |
Token ID | 1037 |
Feature activation | +1.86e+00 |
Pos logit contributions | |
cess | +0.047 |
charged | +0.043 |
aret | +0.042 |
abbit | +0.042 |
eter | +0.042 |
Neg logit contributions | |
done | -0.013 |
wrong | -0.013 |
made | -0.012 |
missing | -0.012 |
had | -0.011 |
Token | and |
Token ID | 290 |
Feature activation | +3.98e+00 |
Pos logit contributions | |
cess | +0.066 |
charged | +0.061 |
eter | +0.061 |
ician | +0.060 |
ymes | +0.060 |
Neg logit contributions | |
made | -0.023 |
all | -0.022 |
on | -0.022 |
done | -0.021 |
wrong | -0.021 |
Token | her |
Token ID | 607 |
Feature activation | +7.91e-01 |
Pos logit contributions | |
cess | +0.117 |
charged | +0.106 |
ck | +0.106 |
aret | +0.104 |
abbit | +0.103 |
Neg logit contributions | |
wrong | -0.064 |
done | -0.063 |
sky | -0.060 |
made | -0.059 |
missing | -0.059 |
Token | mom |
Token ID | 1995 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
cess | +0.020 |
charged | +0.018 |
ck | +0.018 |
aret | +0.017 |
oxy | +0.017 |
Neg logit contributions | |
wrong | -0.017 |
done | -0.017 |
made | -0.017 |
best | -0.017 |
can | -0.017 |
Token | trick |
Token ID | 6908 |
Feature activation | -6.51e-01 |
Pos logit contributions | |
wrong | +0.005 |
done | +0.005 |
made | +0.005 |
sky | +0.005 |
missing | +0.005 |
Neg logit contributions | |
cess | -0.008 |
charged | -0.007 |
ck | -0.007 |
aret | -0.007 |
abbit | -0.007 |
Token | , |
Token ID | 11 |
Feature activation | +2.88e+00 |
Pos logit contributions | |
done | +0.009 |
wrong | +0.009 |
sky | +0.009 |
missing | +0.009 |
made | +0.009 |
Neg logit contributions | |
cess | -0.020 |
ck | -0.018 |
charged | -0.018 |
aret | -0.018 |
abbit | -0.018 |
Token | but |
Token ID | 475 |
Feature activation | +1.07e+00 |
Pos logit contributions | |
cess | +0.068 |
tune | +0.062 |
ck | +0.060 |
eter | +0.059 |
abbit | +0.058 |
Neg logit contributions | |
done | -0.058 |
wrong | -0.056 |
sky | -0.054 |
am | -0.053 |
| -0.053 |
Token | he |
Token ID | 339 |
Feature activation | -2.37e+00 |
Pos logit contributions | |
cess | +0.034 |
ck | +0.031 |
charged | +0.030 |
ishment | +0.030 |
tune | +0.030 |
Neg logit contributions | |
done | -0.013 |
sky | -0.013 |
wrong | -0.012 |
boards | -0.012 |
missing | -0.011 |
Token | didn |
Token ID | 1422 |
Feature activation | -1.91e+00 |
Pos logit contributions | |
wrong | +0.033 |
sky | +0.032 |
done | +0.032 |
missing | +0.030 |
can | +0.029 |
Neg logit contributions | |
cess | -0.074 |
ck | -0.069 |
charged | -0.067 |
oxy | -0.066 |
abbit | -0.066 |
Token | 't |
Token ID | 470 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
done | +0.023 |
wrong | +0.022 |
made | +0.021 |
sky | +0.021 |
had | +0.020 |
Neg logit contributions | |
cess | -0.063 |
ck | -0.058 |
charged | -0.057 |
aret | -0.056 |
oxy | -0.056 |
Token | know |
Token ID | 760 |
Feature activation | -1.25e+00 |
Pos logit contributions | |
cess | +0.031 |
ck | +0.028 |
aret | +0.025 |
tune | +0.025 |
abbit | +0.025 |
Neg logit contributions | |
sky | -0.032 |
wrong | -0.032 |
missing | -0.031 |
done | -0.031 |
boards | -0.030 |
Token | that |
Token ID | 326 |
Feature activation | -8.92e-01 |
Pos logit contributions | |
done | +0.027 |
sky | +0.026 |
wrong | +0.025 |
made | +0.025 |
missing | +0.025 |
Neg logit contributions | |
cess | -0.029 |
ck | -0.027 |
charged | -0.025 |
aret | -0.025 |
abbit | -0.024 |
Token | the |
Token ID | 262 |
Feature activation | -1.74e+00 |
Pos logit contributions | |
done | +0.013 |
wrong | +0.012 |
sky | +0.012 |
missing | +0.011 |
made | +0.011 |
Neg logit contributions | |
cess | -0.027 |
ck | -0.024 |
charged | -0.024 |
aret | -0.024 |
abbit | -0.024 |
Token | naughty |
Token ID | 45128 |
Feature activation | +9.38e-01 |
Pos logit contributions | |
done | +0.036 |
were | +0.035 |
had | +0.035 |
has | +0.034 |
boards | +0.033 |
Neg logit contributions | |
tune | -0.038 |
cess | -0.036 |
tunes | -0.034 |
melody | -0.033 |
ymes | -0.033 |
Token | magician |
Token ID | 42055 |
Feature activation | -8.05e-01 |
Pos logit contributions | |
cess | +0.022 |
tune | +0.020 |
tunes | +0.019 |
ymes | +0.018 |
aret | +0.018 |
Neg logit contributions | |
done | -0.020 |
missing | -0.019 |
wrong | -0.019 |
sky | -0.018 |
has | -0.018 |
Token | His |
Token ID | 2399 |
Feature activation | -9.74e-02 |
Pos logit contributions | |
done | +0.004 |
wrong | +0.004 |
sky | +0.004 |
made | +0.004 |
had | +0.004 |
Neg logit contributions | |
cess | -0.018 |
charged | -0.017 |
ck | -0.017 |
aret | -0.017 |
abbit | -0.017 |
Token | name |
Token ID | 1438 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
wrong | +0.001 |
done | +0.001 |
made | +0.001 |
had | +0.001 |
sky | +0.001 |
Neg logit contributions | |
cess | -0.003 |
oxy | -0.003 |
abbit | -0.003 |
charged | -0.003 |
aret | -0.003 |
Token | was |
Token ID | 373 |
Feature activation | -4.86e-01 |
Pos logit contributions | |
wrong | +0.027 |
done | +0.027 |
missing | +0.024 |
best | +0.022 |
sky | +0.021 |
Neg logit contributions | |
cess | -0.070 |
oxy | -0.065 |
ck | -0.064 |
chick | -0.062 |
cle | -0.061 |
Token | Mr |
Token ID | 1770 |
Feature activation | +4.04e+00 |
Pos logit contributions | |
done | +0.011 |
wrong | +0.011 |
missing | +0.011 |
sky | +0.011 |
can | +0.010 |
Neg logit contributions | |
cess | -0.011 |
oxy | -0.009 |
abbit | -0.009 |
charged | -0.009 |
aret | -0.009 |
Token | . |
Token ID | 13 |
Feature activation | -2.07e-01 |
Pos logit contributions | |
cess | +0.130 |
charged | +0.124 |
ck | +0.121 |
aret | +0.120 |
rily | +0.116 |
Neg logit contributions | |
wrong | -0.055 |
had | -0.053 |
made | -0.053 |
can | -0.052 |
sky | -0.052 |
Token | Sunshine |
Token ID | 32210 |
Feature activation | +4.17e+00 |
Pos logit contributions | |
wrong | +0.005 |
done | +0.005 |
made | +0.005 |
missing | +0.005 |
sky | +0.005 |
Neg logit contributions | |
cess | -0.004 |
charged | -0.004 |
ck | -0.004 |
abbit | -0.004 |
aret | -0.004 |
Token | . |
Token ID | 13 |
Feature activation | +8.97e-02 |
Pos logit contributions | |
cess | +0.129 |
charged | +0.120 |
ck | +0.118 |
aret | +0.117 |
ishment | +0.115 |
Neg logit contributions | |
wrong | -0.061 |
made | -0.059 |
had | -0.059 |
done | -0.059 |
can | -0.057 |
Token | He |
Token ID | 679 |
Feature activation | -9.51e-01 |
Pos logit contributions | |
cess | +0.003 |
charged | +0.002 |
aret | +0.002 |
abbit | +0.002 |
ck | +0.002 |
Neg logit contributions | |
done | -0.001 |
wrong | -0.001 |
missing | -0.001 |
made | -0.001 |
can | -0.001 |
Token | was |
Token ID | 373 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
wrong | +0.010 |
done | +0.010 |
sky | +0.009 |
best | +0.008 |
missing | +0.008 |
Neg logit contributions | |
cess | -0.032 |
oxy | -0.029 |
ck | -0.029 |
abbit | -0.028 |
cle | -0.028 |
Token | carrying |
Token ID | 6872 |
Feature activation | -3.27e-01 |
Pos logit contributions | |
boards | +0.016 |
can | +0.014 |
done | +0.014 |
wrong | +0.014 |
sky | +0.014 |
Neg logit contributions | |
cess | -0.035 |
abbit | -0.031 |
aret | -0.031 |
ck | -0.031 |
charged | -0.031 |
Token | a |
Token ID | 257 |
Feature activation | -7.62e-01 |
Pos logit contributions | |
done | +0.005 |
wrong | +0.004 |
has | +0.004 |
can | +0.004 |
sky | +0.004 |
Neg logit contributions | |
cess | -0.010 |
ck | -0.009 |
charged | -0.009 |
abbit | -0.009 |
oxy | -0.009 |
Token | far |
Token ID | 1290 |
Feature activation | +1.79e+00 |
Pos logit contributions | |
done | +0.017 |
had | +0.016 |
has | +0.016 |
wrong | +0.015 |
can | +0.015 |
Neg logit contributions | |
cess | -0.030 |
chick | -0.028 |
aret | -0.028 |
tune | -0.028 |
charged | -0.028 |
Token | my |
Token ID | 1820 |
Feature activation | -1.31e-01 |
Pos logit contributions | |
cess | +0.045 |
charged | +0.041 |
ck | +0.040 |
aret | +0.040 |
abbit | +0.039 |
Neg logit contributions | |
done | -0.035 |
wrong | -0.034 |
missing | -0.033 |
made | -0.033 |
had | -0.032 |
Token | ard |
Token ID | 446 |
Feature activation | +5.50e-01 |
Pos logit contributions | |
done | +0.002 |
wrong | +0.002 |
am | +0.002 |
missing | +0.002 |
best | +0.002 |
Neg logit contributions | |
cess | -0.003 |
chick | -0.003 |
tune | -0.003 |
ck | -0.003 |
abbit | -0.003 |
Token | when |
Token ID | 618 |
Feature activation | -7.58e-01 |
Pos logit contributions | |
cess | +0.012 |
charged | +0.011 |
cker | +0.010 |
oing | +0.010 |
ishment | +0.010 |
Neg logit contributions | |
done | -0.011 |
wrong | -0.010 |
Everything | -0.010 |
has | -0.010 |
am | -0.010 |
Token | he |
Token ID | 339 |
Feature activation | -2.99e+00 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.009 |
can | +0.009 |
has | +0.008 |
made | +0.008 |
Neg logit contributions | |
cess | -0.022 |
God | -0.020 |
charged | -0.020 |
cle | -0.019 |
ck | -0.019 |
Token | came |
Token ID | 1625 |
Feature activation | -1.48e+00 |
Pos logit contributions | |
wrong | +0.041 |
best | +0.036 |
done | +0.036 |
missing | +0.033 |
biggest | +0.033 |
Neg logit contributions | |
cess | -0.088 |
ck | -0.084 |
oxy | -0.080 |
charged | -0.077 |
feat | -0.077 |
Token | across |
Token ID | 1973 |
Feature activation | -2.24e+00 |
Pos logit contributions | |
done | +0.019 |
boards | +0.018 |
wrong | +0.018 |
can | +0.016 |
toilet | +0.016 |
Neg logit contributions | |
cess | -0.046 |
abbit | -0.042 |
charged | -0.042 |
ck | -0.041 |
eter | -0.041 |
Token | a |
Token ID | 257 |
Feature activation | -1.09e-01 |
Pos logit contributions | |
done | +0.033 |
wrong | +0.031 |
best | +0.029 |
has | +0.029 |
can | +0.028 |
Neg logit contributions | |
cess | -0.063 |
oxy | -0.059 |
chick | -0.059 |
abbit | -0.058 |
ck | -0.058 |
Token | tall |
Token ID | 7331 |
Feature activation | +2.87e+00 |
Pos logit contributions | |
done | +0.002 |
made | +0.001 |
wrong | +0.001 |
has | +0.001 |
had | +0.001 |
Neg logit contributions | |
cess | -0.003 |
chick | -0.003 |
tune | -0.003 |
abbit | -0.003 |
feat | -0.003 |
Token | stack |
Token ID | 8931 |
Feature activation | -1.64e+00 |
Pos logit contributions | |
cess | +0.075 |
charged | +0.066 |
ck | +0.066 |
abbit | +0.066 |
aret | +0.066 |
Neg logit contributions | |
done | -0.054 |
wrong | -0.052 |
missing | -0.050 |
made | -0.049 |
best | -0.048 |
Token | of |
Token ID | 286 |
Feature activation | +9.09e-01 |
Pos logit contributions | |
done | +0.025 |
wrong | +0.025 |
can | +0.023 |
missing | +0.023 |
best | +0.023 |
Neg logit contributions | |
cess | -0.045 |
ck | -0.043 |
charged | -0.041 |
rily | -0.041 |
aret | -0.041 |
Token | and |
Token ID | 290 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
wrong | +0.005 |
done | +0.005 |
sky | +0.005 |
made | +0.005 |
missing | +0.005 |
Neg logit contributions | |
cess | -0.010 |
ck | -0.009 |
charged | -0.009 |
aret | -0.009 |
abbit | -0.009 |
Token | said |
Token ID | 531 |
Feature activation | -1.97e+00 |
Pos logit contributions | |
cess | +0.067 |
charged | +0.061 |
ck | +0.060 |
aret | +0.060 |
abbit | +0.059 |
Neg logit contributions | |
wrong | -0.035 |
done | -0.034 |
sky | -0.032 |
missing | -0.032 |
made | -0.032 |
Token | , |
Token ID | 11 |
Feature activation | +1.72e-01 |
Pos logit contributions | |
done | +0.025 |
wrong | +0.025 |
sky | +0.025 |
missing | +0.024 |
made | +0.023 |
Neg logit contributions | |
cess | -0.063 |
charged | -0.057 |
ck | -0.057 |
aret | -0.056 |
abbit | -0.055 |
Token | " |
Token ID | 366 |
Feature activation | +2.58e+00 |
Pos logit contributions | |
cess | +0.006 |
charged | +0.005 |
ck | +0.005 |
aret | +0.005 |
abbit | +0.005 |
Neg logit contributions | |
wrong | -0.002 |
done | -0.002 |
made | -0.002 |
sky | -0.002 |
missing | -0.002 |
Token | Me |
Token ID | 5308 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
cess | +0.077 |
charged | +0.071 |
ck | +0.070 |
aret | +0.069 |
abbit | +0.068 |
Neg logit contributions | |
made | -0.040 |
done | -0.039 |
had | -0.039 |
sky | -0.039 |
wrong | -0.039 |
Token | too |
Token ID | 1165 |
Feature activation | -6.07e-01 |
Pos logit contributions | |
cess | +0.058 |
charged | +0.054 |
ck | +0.053 |
aret | +0.052 |
abbit | +0.052 |
Neg logit contributions | |
wrong | -0.022 |
done | -0.021 |
sky | -0.020 |
made | -0.020 |
missing | -0.020 |
Token | , |
Token ID | 11 |
Feature activation | +2.24e+00 |
Pos logit contributions | |
wrong | +0.010 |
done | +0.010 |
made | +0.010 |
sky | +0.010 |
missing | +0.010 |
Neg logit contributions | |
cess | -0.017 |
charged | -0.016 |
ck | -0.015 |
aret | -0.015 |
abbit | -0.015 |
Token | Lily |
Token ID | 20037 |
Feature activation | +2.82e+00 |
Pos logit contributions | |
cess | +0.057 |
charged | +0.051 |
ck | +0.050 |
aret | +0.049 |
abbit | +0.049 |
Neg logit contributions | |
wrong | -0.045 |
done | -0.045 |
made | -0.043 |
sky | -0.043 |
missing | -0.042 |
Token | . |
Token ID | 13 |
Feature activation | +1.67e-01 |
Pos logit contributions | |
cess | +0.087 |
ck | +0.082 |
charged | +0.081 |
oxy | +0.079 |
aret | +0.079 |
Neg logit contributions | |
made | -0.042 |
wrong | -0.041 |
done | -0.041 |
had | -0.039 |
can | -0.039 |
Token | Cake |
Token ID | 30799 |
Feature activation | +6.00e-01 |
Pos logit contributions | |
cess | +0.005 |
ck | +0.004 |
charged | +0.004 |
aret | +0.004 |
abbit | +0.004 |
Neg logit contributions | |
done | -0.003 |
wrong | -0.003 |
made | -0.003 |
sky | -0.003 |
can | -0.003 |
Token | is |
Token ID | 318 |
Feature activation | -1.37e-01 |
Pos logit contributions | |
cess | +0.020 |
charged | +0.019 |
ck | +0.018 |
aret | +0.018 |
oxy | +0.018 |
Neg logit contributions | |
wrong | -0.007 |
missing | -0.006 |
done | -0.006 |
sky | -0.006 |
made | -0.005 |
Token | and |
Token ID | 290 |
Feature activation | +1.76e+00 |
Pos logit contributions | |
cess | +0.014 |
charged | +0.013 |
aret | +0.013 |
ck | +0.013 |
abbit | +0.013 |
Neg logit contributions | |
done | -0.005 |
wrong | -0.005 |
made | -0.005 |
on | -0.005 |
sky | -0.005 |
Token | pretended |
Token ID | 36766 |
Feature activation | -1.77e+00 |
Pos logit contributions | |
cess | +0.049 |
ck | +0.045 |
charged | +0.045 |
aret | +0.044 |
oxy | +0.043 |
Neg logit contributions | |
wrong | -0.030 |
done | -0.029 |
sky | -0.028 |
missing | -0.028 |
made | -0.027 |
Token | it |
Token ID | 340 |
Feature activation | -5.34e+00 |
Pos logit contributions | |
done | +0.026 |
wrong | +0.026 |
sky | +0.026 |
missing | +0.025 |
can | +0.024 |
Neg logit contributions | |
cess | -0.049 |
ck | -0.045 |
aret | -0.043 |
oxy | -0.043 |
eter | -0.042 |
Token | was |
Token ID | 373 |
Feature activation | -1.79e+00 |
Pos logit contributions | |
wrong | +0.058 |
done | +0.054 |
sky | +0.048 |
missing | +0.047 |
best | +0.045 |
Neg logit contributions | |
ck | -0.166 |
cess | -0.162 |
chick | -0.155 |
ymes | -0.153 |
oxy | -0.152 |
Token | tea |
Token ID | 8887 |
Feature activation | -8.90e-01 |
Pos logit contributions | |
wrong | +0.023 |
done | +0.023 |
sky | +0.023 |
can | +0.022 |
has | +0.022 |
Neg logit contributions | |
cess | -0.053 |
ck | -0.049 |
aret | -0.048 |
chick | -0.048 |
charged | -0.048 |
Token | . |
Token ID | 13 |
Feature activation | -4.65e-01 |
Pos logit contributions | |
wrong | +0.016 |
done | +0.016 |
missing | +0.015 |
sky | +0.015 |
can | +0.014 |
Neg logit contributions | |
cess | -0.024 |
ck | -0.021 |
charged | -0.021 |
oxy | -0.021 |
aret | -0.021 |
Token | They |
Token ID | 1119 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
wrong | +0.008 |
done | +0.008 |
made | +0.008 |
sky | +0.008 |
missing | +0.008 |
Neg logit contributions | |
cess | -0.013 |
charged | -0.012 |
ck | -0.012 |
aret | -0.012 |
abbit | -0.011 |
Token | enjoyed |
Token ID | 8359 |
Feature activation | +8.43e-01 |
Pos logit contributions | |
wrong | +0.018 |
done | +0.017 |
missing | +0.016 |
sky | +0.016 |
best | +0.016 |
Neg logit contributions | |
cess | -0.034 |
ck | -0.032 |
charged | -0.032 |
aret | -0.031 |
oxy | -0.031 |
Token | their |
Token ID | 511 |
Feature activation | -7.46e-01 |
Pos logit contributions | |
cess | +0.025 |
charged | +0.023 |
ck | +0.023 |
aret | +0.022 |
abbit | +0.022 |
Neg logit contributions | |
wrong | -0.013 |
done | -0.013 |
made | -0.013 |
sky | -0.012 |
missing | -0.012 |
Token | tea |
Token ID | 8887 |
Feature activation | -6.19e-01 |
Pos logit contributions | |
wrong | +0.011 |
done | +0.011 |
made | +0.010 |
sky | +0.010 |
missing | +0.010 |
Neg logit contributions | |
cess | -0.022 |
charged | -0.021 |
ck | -0.020 |
aret | -0.020 |
abbit | -0.020 |
Token | party |
Token ID | 2151 |
Feature activation | +8.41e-01 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.010 |
made | +0.009 |
sky | +0.009 |
missing | +0.009 |
Neg logit contributions | |
cess | -0.018 |
charged | -0.017 |
ck | -0.017 |
aret | -0.016 |
abbit | -0.016 |
Token | He |
Token ID | 679 |
Feature activation | -2.20e+00 |
Pos logit contributions | |
done | +0.007 |
wrong | +0.006 |
missing | +0.006 |
made | +0.006 |
sky | +0.006 |
Neg logit contributions | |
cess | -0.009 |
abbit | -0.008 |
eter | -0.008 |
charged | -0.008 |
ck | -0.008 |
Token | sw |
Token ID | 1509 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
wrong | +0.033 |
done | +0.031 |
sky | +0.030 |
missing | +0.030 |
best | +0.027 |
Neg logit contributions | |
cess | -0.066 |
ck | -0.064 |
charged | -0.060 |
abbit | -0.059 |
aret | -0.058 |
Token | am |
Token ID | 321 |
Feature activation | -1.36e+00 |
Pos logit contributions | |
cess | +0.025 |
charged | +0.022 |
ck | +0.022 |
aret | +0.021 |
abbit | +0.021 |
Neg logit contributions | |
wrong | -0.039 |
done | -0.038 |
sky | -0.037 |
made | -0.037 |
missing | -0.037 |
Token | and |
Token ID | 290 |
Feature activation | +6.63e-01 |
Pos logit contributions | |
done | +0.025 |
sky | +0.025 |
wrong | +0.025 |
missing | +0.024 |
has | +0.023 |
Neg logit contributions | |
cess | -0.033 |
ck | -0.032 |
rily | -0.030 |
tune | -0.029 |
aret | -0.029 |
Token | sw |
Token ID | 1509 |
Feature activation | +1.16e+00 |
Pos logit contributions | |
cess | +0.018 |
ck | +0.017 |
charged | +0.016 |
abbit | +0.016 |
tune | +0.016 |
Neg logit contributions | |
wrong | -0.012 |
done | -0.011 |
sky | -0.011 |
missing | -0.011 |
best | -0.010 |
Token | am |
Token ID | 321 |
Feature activation | -2.04e+00 |
Pos logit contributions | |
cess | +0.021 |
ck | +0.018 |
charged | +0.018 |
aret | +0.017 |
abbit | +0.017 |
Neg logit contributions | |
wrong | -0.033 |
done | -0.032 |
sky | -0.032 |
missing | -0.031 |
made | -0.031 |
Token | until |
Token ID | 1566 |
Feature activation | -2.46e+00 |
Pos logit contributions | |
done | +0.042 |
sky | +0.041 |
wrong | +0.040 |
missing | +0.040 |
has | +0.039 |
Neg logit contributions | |
cess | -0.044 |
ck | -0.043 |
tune | -0.040 |
charged | -0.039 |
rily | -0.039 |
Token | he |
Token ID | 339 |
Feature activation | -4.42e+00 |
Pos logit contributions | |
boards | +0.034 |
has | +0.033 |
done | +0.033 |
living | +0.033 |
wrong | +0.032 |
Neg logit contributions | |
ck | -0.067 |
cess | -0.065 |
feat | -0.059 |
chick | -0.058 |
swat | -0.057 |
Token | reached |
Token ID | 4251 |
Feature activation | -1.89e+00 |
Pos logit contributions | |
wrong | +0.056 |
sky | +0.048 |
missing | +0.048 |
done | +0.046 |
best | +0.044 |
Neg logit contributions | |
cess | -0.142 |
ck | -0.135 |
oxy | -0.124 |
feat | -0.121 |
charged | -0.121 |
Token | a |
Token ID | 257 |
Feature activation | -2.08e+00 |
Pos logit contributions | |
boards | +0.040 |
done | +0.037 |
wrong | +0.036 |
sky | +0.035 |
can | +0.034 |
Neg logit contributions | |
cess | -0.041 |
ck | -0.038 |
feat | -0.037 |
rily | -0.036 |
eter | -0.035 |
Token | hidden |
Token ID | 7104 |
Feature activation | -2.58e-01 |
Pos logit contributions | |
done | +0.034 |
boards | +0.031 |
ained | +0.029 |
ed | +0.029 |
best | +0.028 |
Neg logit contributions | |
tune | -0.054 |
cess | -0.054 |
chick | -0.053 |
feat | -0.050 |
abbit | -0.049 |
Token | to |
Token ID | 284 |
Feature activation | +3.18e+00 |
Pos logit contributions | |
sky | +0.020 |
done | +0.019 |
| +0.019 |
boards | +0.018 |
has | +0.018 |
Neg logit contributions | |
cess | -0.030 |
ck | -0.026 |
aret | -0.026 |
eter | -0.025 |
charged | -0.025 |
Token | always |
Token ID | 1464 |
Feature activation | +2.37e+00 |
Pos logit contributions | |
cess | +0.085 |
ck | +0.079 |
charged | +0.075 |
aret | +0.075 |
tune | +0.073 |
Neg logit contributions | |
wrong | -0.059 |
sky | -0.058 |
boards | -0.057 |
missing | -0.054 |
| -0.054 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.32e+00 |
Pos logit contributions | |
cess | +0.052 |
ck | +0.048 |
tune | +0.045 |
charged | +0.045 |
feat | +0.043 |
Neg logit contributions | |
sky | -0.056 |
boards | -0.054 |
wrong | -0.052 |
| -0.051 |
missing | -0.050 |
Token | his |
Token ID | 465 |
Feature activation | -5.58e-01 |
Pos logit contributions | |
cess | +0.034 |
ck | +0.032 |
charged | +0.032 |
aret | +0.031 |
abbit | +0.030 |
Neg logit contributions | |
sky | -0.025 |
done | -0.023 |
can | -0.023 |
wrong | -0.023 |
has | -0.023 |
Token | little |
Token ID | 1310 |
Feature activation | +9.06e-01 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.010 |
made | +0.010 |
sky | +0.010 |
had | +0.010 |
Neg logit contributions | |
cess | -0.014 |
charged | -0.013 |
abbit | -0.013 |
ishment | -0.013 |
tune | -0.012 |
Token | basketball |
Token ID | 9669 |
Feature activation | +3.77e-02 |
Pos logit contributions | |
cess | +0.022 |
charged | +0.021 |
abbit | +0.021 |
eter | +0.020 |
ishment | +0.020 |
Neg logit contributions | |
done | -0.018 |
wrong | -0.017 |
made | -0.016 |
sky | -0.016 |
had | -0.016 |
Token | close |
Token ID | 1969 |
Feature activation | +7.06e-01 |
Pos logit contributions | |
cess | +0.001 |
charged | +0.001 |
ck | +0.001 |
aret | +0.001 |
abbit | +0.001 |
Neg logit contributions | |
done | -0.001 |
wrong | -0.001 |
sky | -0.001 |
made | -0.001 |
can | -0.001 |
Token | to |
Token ID | 284 |
Feature activation | +1.83e+00 |
Pos logit contributions | |
cess | +0.020 |
charged | +0.019 |
eter | +0.018 |
aret | +0.018 |
ck | +0.018 |
Neg logit contributions | |
done | -0.011 |
wrong | -0.011 |
sky | -0.010 |
made | -0.010 |
missing | -0.010 |
Token | him |
Token ID | 683 |
Feature activation | +7.19e-01 |
Pos logit contributions | |
cess | +0.050 |
charged | +0.045 |
ck | +0.045 |
aret | +0.044 |
abbit | +0.043 |
Neg logit contributions | |
done | -0.032 |
wrong | -0.032 |
sky | -0.031 |
made | -0.030 |
missing | -0.030 |
Token | . |
Token ID | 13 |
Feature activation | -3.51e-02 |
Pos logit contributions | |
cess | +0.022 |
charged | +0.020 |
ck | +0.020 |
aret | +0.020 |
oxy | +0.019 |
Neg logit contributions | |
wrong | -0.011 |
done | -0.011 |
made | -0.010 |
missing | -0.010 |
can | -0.010 |
Token | The |
Token ID | 383 |
Feature activation | -1.52e+00 |
Pos logit contributions | |
done | +0.000 |
wrong | +0.000 |
sky | +0.000 |
made | +0.000 |
can | +0.000 |
Neg logit contributions | |
cess | -0.001 |
ck | -0.001 |
charged | -0.001 |
aret | -0.001 |
oxy | -0.001 |
Token | saw |
Token ID | 2497 |
Feature activation | -2.01e+00 |
Pos logit contributions | |
wrong | +0.029 |
done | +0.027 |
sky | +0.025 |
missing | +0.024 |
best | +0.023 |
Neg logit contributions | |
cess | -0.073 |
ck | -0.068 |
charged | -0.066 |
abbit | -0.064 |
chick | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -7.82e-01 |
Pos logit contributions | |
done | +0.030 |
has | +0.026 |
can | +0.025 |
wrong | +0.025 |
sky | +0.024 |
Neg logit contributions | |
cess | -0.060 |
ck | -0.057 |
chick | -0.055 |
charged | -0.055 |
feat | -0.054 |
Token | fox |
Token ID | 21831 |
Feature activation | +1.22e+00 |
Pos logit contributions | |
done | +0.020 |
boards | +0.018 |
has | +0.017 |
wrong | +0.017 |
Everything | +0.017 |
Neg logit contributions | |
tune | -0.014 |
chick | -0.014 |
cess | -0.013 |
feat | -0.013 |
chicks | -0.013 |
Token | again |
Token ID | 757 |
Feature activation | +3.27e-01 |
Pos logit contributions | |
cess | +0.034 |
ck | +0.032 |
charged | +0.031 |
abbit | +0.030 |
oing | +0.030 |
Neg logit contributions | |
done | -0.020 |
wrong | -0.019 |
sky | -0.019 |
made | -0.018 |
can | -0.018 |
Token | and |
Token ID | 290 |
Feature activation | +2.68e+00 |
Pos logit contributions | |
cess | +0.009 |
ck | +0.008 |
charged | +0.008 |
abbit | +0.007 |
oxy | +0.007 |
Neg logit contributions | |
wrong | -0.006 |
done | -0.006 |
sky | -0.006 |
missing | -0.006 |
can | -0.006 |
Token | this |
Token ID | 428 |
Feature activation | -3.84e+00 |
Pos logit contributions | |
cess | +0.082 |
ck | +0.076 |
charged | +0.072 |
chick | +0.071 |
aret | +0.070 |
Neg logit contributions | |
done | -0.039 |
wrong | -0.038 |
sky | -0.036 |
missing | -0.035 |
best | -0.035 |
Token | time |
Token ID | 640 |
Feature activation | -2.51e+00 |
Pos logit contributions | |
wrong | +0.050 |
can | +0.048 |
done | +0.048 |
boards | +0.048 |
missing | +0.048 |
Neg logit contributions | |
ck | -0.105 |
tune | -0.101 |
feat | -0.099 |
oxy | -0.098 |
cess | -0.098 |
Token | she |
Token ID | 673 |
Feature activation | -2.27e+00 |
Pos logit contributions | |
done | +0.036 |
wrong | +0.034 |
sky | +0.034 |
am | +0.032 |
best | +0.032 |
Neg logit contributions | |
cess | -0.071 |
ck | -0.064 |
feat | -0.062 |
eter | -0.062 |
oxy | -0.062 |
Token | was |
Token ID | 373 |
Feature activation | -2.00e+00 |
Pos logit contributions | |
wrong | +0.028 |
done | +0.025 |
sky | +0.024 |
missing | +0.023 |
best | +0.023 |
Neg logit contributions | |
cess | -0.074 |
ck | -0.071 |
charged | -0.068 |
oxy | -0.066 |
feat | -0.066 |
Token | able |
Token ID | 1498 |
Feature activation | -2.56e+00 |
Pos logit contributions | |
can | +0.033 |
sky | +0.032 |
wrong | +0.031 |
boards | +0.031 |
has | +0.030 |
Neg logit contributions | |
cess | -0.056 |
ck | -0.052 |
charged | -0.050 |
aret | -0.048 |
rily | -0.047 |
Token | to |
Token ID | 284 |
Feature activation | +2.44e+00 |
Pos logit contributions | |
sky | +0.037 |
wrong | +0.036 |
done | +0.036 |
can | +0.035 |
made | +0.034 |
Neg logit contributions | |
cess | -0.074 |
ck | -0.068 |
charged | -0.068 |
abbit | -0.067 |
eter | -0.066 |
Token | animals |
Token ID | 4695 |
Feature activation | -9.53e-01 |
Pos logit contributions | |
done | +0.020 |
wrong | +0.020 |
can | +0.019 |
missing | +0.019 |
sky | +0.018 |
Neg logit contributions | |
cess | -0.019 |
abbit | -0.017 |
charged | -0.017 |
ck | -0.017 |
chick | -0.017 |
Token | escaped |
Token ID | 13537 |
Feature activation | -7.76e-02 |
Pos logit contributions | |
wrong | +0.013 |
sky | +0.012 |
missing | +0.011 |
best | +0.011 |
biggest | +0.011 |
Neg logit contributions | |
cess | -0.028 |
ck | -0.028 |
chick | -0.026 |
chicks | -0.025 |
charged | -0.025 |
Token | the |
Token ID | 262 |
Feature activation | -1.21e+00 |
Pos logit contributions | |
done | +0.002 |
wrong | +0.001 |
has | +0.001 |
sky | +0.001 |
best | +0.001 |
Neg logit contributions | |
cess | -0.002 |
rily | -0.002 |
ck | -0.002 |
aret | -0.002 |
charged | -0.002 |
Token | barn |
Token ID | 25203 |
Feature activation | -1.58e+00 |
Pos logit contributions | |
done | +0.025 |
wrong | +0.023 |
made | +0.022 |
can | +0.022 |
am | +0.022 |
Neg logit contributions | |
cess | -0.028 |
chick | -0.027 |
charged | -0.027 |
aret | -0.026 |
chicks | -0.026 |
Token | before |
Token ID | 878 |
Feature activation | -1.43e+00 |
Pos logit contributions | |
wrong | +0.028 |
done | +0.027 |
can | +0.026 |
am | +0.026 |
missing | +0.026 |
Neg logit contributions | |
cess | -0.040 |
charged | -0.035 |
eter | -0.035 |
rily | -0.035 |
aret | -0.035 |
Token | it |
Token ID | 340 |
Feature activation | -4.60e+00 |
Pos logit contributions | |
done | +0.024 |
wrong | +0.024 |
can | +0.022 |
has | +0.022 |
made | +0.022 |
Neg logit contributions | |
cess | -0.039 |
ck | -0.035 |
oxy | -0.034 |
abbit | -0.034 |
charged | -0.034 |
Token | was |
Token ID | 373 |
Feature activation | -2.33e+00 |
Pos logit contributions | |
wrong | +0.052 |
done | +0.048 |
sky | +0.047 |
missing | +0.047 |
best | +0.045 |
Neg logit contributions | |
cess | -0.143 |
ck | -0.141 |
chick | -0.132 |
elope | -0.131 |
aret | -0.131 |
Token | completely |
Token ID | 3190 |
Feature activation | -1.98e+00 |
Pos logit contributions | |
sky | +0.034 |
boards | +0.034 |
can | +0.034 |
has | +0.033 |
am | +0.033 |
Neg logit contributions | |
cess | -0.064 |
ck | -0.063 |
aret | -0.059 |
charged | -0.059 |
rily | -0.058 |
Token | destroyed |
Token ID | 6572 |
Feature activation | +2.80e-01 |
Pos logit contributions | |
can | +0.032 |
sky | +0.032 |
has | +0.031 |
boards | +0.030 |
am | +0.030 |
Neg logit contributions | |
cess | -0.050 |
ck | -0.050 |
charged | -0.049 |
eter | -0.047 |
rily | -0.046 |
Token | by |
Token ID | 416 |
Feature activation | +1.29e+00 |
Pos logit contributions | |
cess | +0.008 |
charged | +0.007 |
ck | +0.007 |
abbit | +0.007 |
aret | +0.007 |
Neg logit contributions | |
done | -0.005 |
wrong | -0.005 |
sky | -0.005 |
made | -0.005 |
can | -0.004 |
Token | the |
Token ID | 262 |
Feature activation | -8.78e-01 |
Pos logit contributions | |
cess | +0.037 |
charged | +0.034 |
ck | +0.034 |
abbit | +0.033 |
aret | +0.033 |
Neg logit contributions | |
done | -0.020 |
wrong | -0.020 |
sky | -0.020 |
can | -0.019 |
made | -0.019 |
Token | the |
Token ID | 262 |
Feature activation | -1.71e-01 |
Pos logit contributions | |
cess | +0.082 |
ck | +0.074 |
aret | +0.074 |
charged | +0.073 |
abbit | +0.071 |
Neg logit contributions | |
done | -0.050 |
wrong | -0.049 |
sky | -0.047 |
missing | -0.047 |
made | -0.046 |
Token | loud |
Token ID | 7812 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
done | +0.004 |
wrong | +0.004 |
made | +0.004 |
can | +0.003 |
missing | +0.003 |
Neg logit contributions | |
cess | -0.004 |
charged | -0.003 |
tune | -0.003 |
aret | -0.003 |
ck | -0.003 |
Token | noises |
Token ID | 26782 |
Feature activation | +1.25e-01 |
Pos logit contributions | |
cess | +0.037 |
charged | +0.033 |
aret | +0.032 |
ck | +0.032 |
tune | +0.031 |
Neg logit contributions | |
done | -0.035 |
wrong | -0.034 |
missing | -0.033 |
made | -0.033 |
sky | -0.032 |
Token | . |
Token ID | 13 |
Feature activation | -8.39e-01 |
Pos logit contributions | |
cess | +0.003 |
ck | +0.003 |
tune | +0.003 |
rily | +0.003 |
charged | +0.003 |
Neg logit contributions | |
done | -0.002 |
wrong | -0.002 |
sky | -0.002 |
missing | -0.002 |
has | -0.002 |
Token | Suddenly |
Token ID | 24975 |
Feature activation | -3.41e+00 |
Pos logit contributions | |
done | +0.014 |
wrong | +0.014 |
made | +0.013 |
missing | +0.013 |
had | +0.013 |
Neg logit contributions | |
cess | -0.024 |
abbit | -0.021 |
charged | -0.021 |
oxy | -0.021 |
eter | -0.021 |
Token | there |
Token ID | 612 |
Feature activation | -4.14e+00 |
Pos logit contributions | |
done | +0.053 |
missing | +0.052 |
wrong | +0.051 |
sky | +0.048 |
has | +0.046 |
Neg logit contributions | |
cess | -0.093 |
ck | -0.085 |
chicks | -0.083 |
chick | -0.082 |
abbit | -0.081 |
Token | was |
Token ID | 373 |
Feature activation | -1.81e+00 |
Pos logit contributions | |
done | +0.045 |
wrong | +0.045 |
missing | +0.042 |
has | +0.037 |
sky | +0.034 |
Neg logit contributions | |
cess | -0.131 |
ck | -0.124 |
rily | -0.120 |
chick | -0.119 |
eter | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | +1.98e-01 |
Pos logit contributions | |
done | +0.027 |
has | +0.025 |
can | +0.025 |
am | +0.025 |
wrong | +0.024 |
Neg logit contributions | |
cess | -0.049 |
eter | -0.046 |
chicks | -0.045 |
chick | -0.045 |
oxy | -0.044 |
Token | knock |
Token ID | 10643 |
Feature activation | -2.39e+00 |
Pos logit contributions | |
cess | +0.005 |
tune | +0.005 |
chick | +0.005 |
abbit | +0.005 |
oing | +0.005 |
Neg logit contributions | |
done | -0.003 |
wrong | -0.003 |
made | -0.003 |
missing | -0.003 |
on | -0.003 |
Token | on |
Token ID | 319 |
Feature activation | -8.48e-02 |
Pos logit contributions | |
done | +0.027 |
wrong | +0.027 |
sky | +0.027 |
missing | +0.027 |
best | +0.026 |
Neg logit contributions | |
cess | -0.071 |
ck | -0.069 |
charged | -0.065 |
oxy | -0.065 |
rily | -0.064 |
Token | the |
Token ID | 262 |
Feature activation | -3.69e+00 |
Pos logit contributions | |
done | +0.001 |
wrong | +0.001 |
missing | +0.001 |
made | +0.001 |
best | +0.001 |
Neg logit contributions | |
cess | -0.002 |
ck | -0.002 |
aret | -0.002 |
charged | -0.002 |
abbit | -0.002 |
Token | keep |
Token ID | 1394 |
Feature activation | +1.81e+00 |
Pos logit contributions | |
cess | +0.068 |
ck | +0.066 |
charged | +0.062 |
feat | +0.058 |
tune | +0.058 |
Neg logit contributions | |
sky | -0.058 |
boards | -0.057 |
wrong | -0.056 |
has | -0.055 |
missing | -0.054 |
Token | her |
Token ID | 607 |
Feature activation | +1.11e+00 |
Pos logit contributions | |
cess | +0.045 |
charged | +0.040 |
ck | +0.040 |
aret | +0.039 |
abbit | +0.038 |
Neg logit contributions | |
wrong | -0.037 |
done | -0.037 |
sky | -0.036 |
made | -0.036 |
missing | -0.035 |
Token | safe |
Token ID | 3338 |
Feature activation | +1.40e+00 |
Pos logit contributions | |
cess | +0.032 |
ck | +0.028 |
oxy | +0.028 |
charged | +0.028 |
aret | +0.028 |
Neg logit contributions | |
wrong | -0.020 |
done | -0.020 |
best | -0.019 |
made | -0.019 |
missing | -0.019 |
Token | . |
Token ID | 13 |
Feature activation | -3.76e-01 |
Pos logit contributions | |
cess | +0.040 |
charged | +0.036 |
ck | +0.036 |
aret | +0.035 |
abbit | +0.035 |
Neg logit contributions | |
wrong | -0.023 |
done | -0.023 |
made | -0.022 |
sky | -0.022 |
missing | -0.021 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -2.77e+00 |
Pos logit contributions | |
wrong | +0.004 |
done | +0.004 |
made | +0.004 |
sky | +0.004 |
can | +0.004 |
Neg logit contributions | |
cess | -0.013 |
ck | -0.012 |
charged | -0.012 |
aret | -0.012 |
abbit | -0.011 |
Token | Once |
Token ID | 7454 |
Feature activation | -4.09e+00 |
Pos logit contributions | |
wrong | +0.044 |
done | +0.044 |
missing | +0.041 |
made | +0.040 |
best | +0.039 |
Neg logit contributions | |
cess | -0.077 |
oxy | -0.070 |
charged | -0.069 |
ck | -0.069 |
abbit | -0.069 |
Token | upon |
Token ID | 2402 |
Feature activation | -1.17e+00 |
Pos logit contributions | |
wrong | +0.048 |
done | +0.047 |
made | +0.045 |
sky | +0.044 |
missing | +0.043 |
Neg logit contributions | |
cess | -0.132 |
aret | -0.123 |
ck | -0.122 |
charged | -0.122 |
abbit | -0.119 |
Token | a |
Token ID | 257 |
Feature activation | -2.13e+00 |
Pos logit contributions | |
wrong | +0.008 |
done | +0.008 |
made | +0.008 |
had | +0.007 |
missing | +0.007 |
Neg logit contributions | |
cess | -0.046 |
charged | -0.043 |
aret | -0.042 |
ck | -0.042 |
abbit | -0.042 |
Token | time |
Token ID | 640 |
Feature activation | -3.61e+00 |
Pos logit contributions | |
done | +0.010 |
wrong | +0.009 |
made | +0.008 |
had | +0.008 |
can | +0.007 |
Neg logit contributions | |
cess | -0.091 |
charged | -0.084 |
aret | -0.082 |
ck | -0.082 |
rily | -0.081 |
Token | , |
Token ID | 11 |
Feature activation | -1.90e+00 |
Pos logit contributions | |
wrong | +0.045 |
done | +0.045 |
sky | +0.043 |
made | +0.042 |
missing | +0.041 |
Neg logit contributions | |
cess | -0.116 |
charged | -0.106 |
ck | -0.105 |
abbit | -0.104 |
aret | -0.104 |
Token | there |
Token ID | 612 |
Feature activation | -4.81e+00 |
Pos logit contributions | |
done | +0.019 |
sky | +0.018 |
wrong | +0.018 |
boards | +0.017 |
missing | +0.017 |
Neg logit contributions | |
cess | -0.061 |
ck | -0.060 |
charged | -0.058 |
abbit | -0.058 |
chick | -0.057 |
Token | became |
Token ID | 2627 |
Feature activation | -6.57e-01 |
Pos logit contributions | |
cess | +0.087 |
ck | +0.080 |
charged | +0.077 |
feat | +0.077 |
aret | +0.077 |
Neg logit contributions | |
wrong | -0.051 |
done | -0.049 |
missing | -0.049 |
sky | -0.048 |
can | -0.045 |
Token | very |
Token ID | 845 |
Feature activation | +1.28e+00 |
Pos logit contributions | |
wrong | +0.008 |
done | +0.008 |
made | +0.008 |
sky | +0.008 |
missing | +0.008 |
Neg logit contributions | |
cess | -0.021 |
charged | -0.020 |
ck | -0.020 |
aret | -0.019 |
abbit | -0.019 |
Token | wise |
Token ID | 10787 |
Feature activation | +2.33e+00 |
Pos logit contributions | |
cess | +0.030 |
ck | +0.027 |
charged | +0.027 |
aret | +0.025 |
abbit | +0.025 |
Neg logit contributions | |
wrong | -0.027 |
done | -0.027 |
sky | -0.026 |
made | -0.026 |
missing | -0.026 |
Token | . |
Token ID | 13 |
Feature activation | -9.11e-01 |
Pos logit contributions | |
cess | +0.065 |
charged | +0.059 |
ck | +0.057 |
abbit | +0.057 |
aret | +0.057 |
Neg logit contributions | |
done | -0.041 |
wrong | -0.040 |
made | -0.038 |
sky | -0.038 |
missing | -0.038 |
Token | <|endoftext|> |
Token ID | 50256 |
Feature activation | -3.00e+00 |
Pos logit contributions | |
sky | +0.009 |
wrong | +0.009 |
done | +0.009 |
can | +0.009 |
had | +0.009 |
Neg logit contributions | |
cess | -0.032 |
charged | -0.030 |
ck | -0.030 |
aret | -0.029 |
oxy | -0.029 |
Token | Once |
Token ID | 7454 |
Feature activation | -3.85e+00 |
Pos logit contributions | |
done | +0.039 |
wrong | +0.039 |
missing | +0.035 |
made | +0.032 |
best | +0.032 |
Neg logit contributions | |
cess | -0.090 |
abbit | -0.083 |
.'" | -0.082 |
oxy | -0.082 |
ymes | -0.081 |
Token | there |
Token ID | 612 |
Feature activation | -3.62e+00 |
Pos logit contributions | |
done | +0.044 |
wrong | +0.044 |
made | +0.042 |
sky | +0.041 |
missing | +0.041 |
Neg logit contributions | |
cess | -0.124 |
aret | -0.116 |
ck | -0.115 |
charged | -0.115 |
abbit | -0.113 |
Token | was |
Token ID | 373 |
Feature activation | -8.70e-01 |
Pos logit contributions | |
wrong | +0.038 |
done | +0.038 |
missing | +0.036 |
sky | +0.035 |
boards | +0.032 |
Neg logit contributions | |
cess | -0.115 |
ck | -0.106 |
charged | -0.105 |
elope | -0.103 |
aret | -0.103 |
Token | an |
Token ID | 281 |
Feature activation | +1.60e+00 |
Pos logit contributions | |
wrong | +0.006 |
done | +0.006 |
made | +0.005 |
sky | +0.005 |
had | +0.005 |
Neg logit contributions | |
cess | -0.035 |
charged | -0.032 |
aret | -0.032 |
ck | -0.031 |
abbit | -0.031 |
Token | ancient |
Token ID | 6156 |
Feature activation | +1.92e+00 |
Pos logit contributions | |
cess | +0.057 |
charged | +0.052 |
rums | +0.051 |
oing | +0.051 |
eter | +0.050 |
Neg logit contributions | |
all | -0.021 |
big | -0.019 |
sky | -0.019 |
best | -0.019 |
amazing | -0.018 |
Token | elephant |
Token ID | 20950 |
Feature activation | +3.24e+00 |
Pos logit contributions | |
cess | +0.050 |
abbit | +0.043 |
charged | +0.043 |
aret | +0.043 |
ck | +0.043 |
Neg logit contributions | |
done | -0.037 |
wrong | -0.036 |
made | -0.034 |
missing | -0.034 |
can | -0.033 |